U.S. patent application number 12/476175 was filed with the patent office on 2009-12-03 for system and method for supporting a virtual appliance.
This patent application is currently assigned to Novell, Inc.. Invention is credited to Peter Bowen, Nathaniel Friedman.
Application Number | 20090300641 12/476175 |
Document ID | / |
Family ID | 41381091 |
Filed Date | 2009-12-03 |
United States Patent
Application |
20090300641 |
Kind Code |
A1 |
Friedman; Nathaniel ; et
al. |
December 3, 2009 |
SYSTEM AND METHOD FOR SUPPORTING A VIRTUAL APPLIANCE
Abstract
A system and method for supporting a virtual appliance is
provided. In particular, a support engine may include an update
server that can manage a workflow to update an appliance in
response to detecting upstream updates to one or more software
components that have been installed for the appliance. For example,
the workflow may generally include managing a rebuild the appliance
to install the upstream updates and further managing an integration
test to verify that the rebuilt appliance behaves correctly with
the upstream updates installed. In addition, the support engine may
further include a support analysis manager that can analyze the
software components that have been installed for the appliance in
view of various heuristic rules to generate a support statement
indicating whether support is available for the appliance.
Inventors: |
Friedman; Nathaniel;
(Munich, DE) ; Bowen; Peter; (Ottawa, CA) |
Correspondence
Address: |
PILLSBURY WINTHROP SHAW PITTMAN, LLP
P.O. BOX 10500
MCLEAN
VA
22102
US
|
Assignee: |
Novell, Inc.
Provo
UT
|
Family ID: |
41381091 |
Appl. No.: |
12/476175 |
Filed: |
June 1, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61057593 |
May 30, 2008 |
|
|
|
Current U.S.
Class: |
718/104 |
Current CPC
Class: |
G06F 11/3664
20130101 |
Class at
Publication: |
718/104 |
International
Class: |
G06F 9/46 20060101
G06F009/46 |
Claims
1. A system for updating a virtual appliance, comprising: an update
server configured to generate a workflow to update one or more
appliances in response to a repository metadata server detecting an
upstream update to one or more origin repositories, wherein the
update server is further configured to: identify one or more
software components associated with the upstream update to the one
or more origin repositories; and identify at least one appliance
having earlier versions of the identified software components
installed therein; a build engine configured to rebuild an image
corresponding to the identified appliance, wherein the rebuilt
image includes the identified software components associated with
the upstream update; and a runtime environment configured to manage
an integration test for the rebuilt image corresponding to the
identified appliance.
2. The system of claim 1, further comprising an appliance
repository configured to store the rebuilt image corresponding to
the identified appliance in response to the integration test
verifying that the rebuilt image functions correctly with the
identified software components associated with the upstream update
installed therein.
3. The system of claim 2, further comprising a development
environment configured to: provide a plurality of public
appliances, wherein the plurality of public appliances include at
least the rebuilt image corresponding to the identified appliance;
and clone the rebuilt image to create a base appliance derived from
the rebuilt image, wherein the base appliance includes the
identified software components associated with the upstream update
installed therein.
4. The system of claim 2, wherein runtime activity that occurs
during the integration test results in one or more changes to the
rebuilt image, wherein the one or more changes are implemented to
accommodate the upstream update.
5. The system of claim 4, wherein the build engine is further
configured to rebuild a second image corresponding to the
identified appliance, wherein the second rebuilt image includes the
one or more changes implemented during the integration test, and
wherein the runtime environment is further configured to manage a
second integration test for the second rebuilt image corresponding
to the identified appliance.
6. The system of claim 1, wherein the update server is further
configured to: generate one or more updates corresponding to the
software components associated with the upstream update installed
for the rebuilt image; and release the updates to one or more
earlier versions of the identified appliance.
7. The system of claim 6, wherein the update server is configured
to release the updates to the earlier versions of the identified
appliance in response to a request received from the earlier
versions of the identified appliance.
8. The system of claim 6, wherein the updates include one or more
access control mechanisms that define one or more criteria
controlling the release to the earlier versions of the identified
appliance.
9. The system of claim 6, wherein the updates include one or more
of a package 6r a post-installation script configured to install
the software components associated with the upstream update to the
earlier versions of the identified appliance.
10. A method for updating a virtual appliance, comprising:
generating a workflow to update one or more appliances at an update
server in response to a repository metadata server detecting an
upstream update to one or more origin repositories; identifying one
or more software components associated with the upstream update to
the one or more origin repositories; identifying at least one
appliance having earlier versions of the identified software
components installed therein; invoking a build engine to rebuild an
image corresponding to the identified appliance, wherein the
rebuilt image includes the identified software components
associated with the upstream update; and invoking a runtime
environment to manage an integration test for the rebuilt image
corresponding to the identified appliance.
11. The method of claim 10, further comprising storing the rebuilt
image corresponding to the identified appliance in an appliance
repository, wherein the rebuilt image is stored in the appliance
repository in response to the integration test verifying that the
rebuilt image functions correctly with the identified software
components associated with the upstream update installed
therein.
12. The method of claim 11, further comprising: providing a
plurality of public appliances, wherein the plurality of public
appliances include at least the rebuilt image corresponding to the
identified appliance; and cloning the rebuilt image to create a
base appliance derived from the rebuilt image, wherein the base
appliance includes the identified software components associated
with the upstream update installed therein.
13. The method of claim 11, wherein runtime activity that occurs
during the integration test results in one or more changes to the
rebuilt image, wherein the one or more changes are implemented to
accommodate the upstream update.
14. The method of claim 13, further comprising: invoking the build
engine to rebuild a second image corresponding to the identified
appliance, wherein the second rebuilt image includes the one or
more changes implemented during the integration test; and invoking
the runtime environment to manage a second integration test for the
second rebuilt image corresponding to the identified appliance.
15. The method of claim 10, further comprising: generating one or
more updates corresponding to the software components associated
with the upstream update installed for the rebuilt image; and
releasing the updates to one or more earlier versions of the
identified appliance.
16. The method of claim 15, wherein the updates are released to the
earlier versions of the identified appliance in response to a
request received from the earlier versions of the identified
appliance.
17. The method of claim 15, wherein the updates include one or more
access control mechanisms that define one or more criteria
controlling the release to the earlier versions of the identified
appliance.
18. The method of claim 15, wherein the updates include one or more
of a package or a post-installation script configured to install
the software components associated with the upstream update to the
earlier versions of the identified appliance.
19. A system for supporting a virtual appliance, comprising: a
support engine configured to analyze supportability and service
support requests for an appliance, wherein the support engine is
further configured to: identify one or more software components
installed for the appliance; analyze the installed software
components in view of one or more rules that express supportability
guidelines for the installed software components; and generate a
support statement indicating whether or not support can be provided
for the installed software components; and a build engine
configured to invoke the support engine in response to a request to
build an image corresponding to the appliance.
20. The system of claim 19, wherein the one or more rules include
heuristics that describe one or more of permitted changes,
permitted modifications, or permitted variants for the installed
software components.
21. The system of claim 19, wherein the support engine is further
configured to: retrieve metadata describing the installed software
components; and analyze the metadata describing the installed
software components in view of the one or more rules, wherein the
one or more rules express the supportability guidelines in terms of
the metadata describing the installed software components.
22. The system of claim 19, wherein the support engine is further
configured to: receive a support request describing a problem
associated with the appliance; and reference the support statement
to determine whether or not support can be provided for the problem
associated with the appliance.
23. The system of claim 22, wherein the support request includes a
snapshot of the appliance in a state associated with the
problem.
24. The system of claim 23, wherein the support engine is further
configured to generate a second support statement indicating
whether or not support can be provided for the appliance in the
problem state.
25. The system of claim 22, wherein the support engine is further
configured to generate a response to the support request in
response to determining that support can be provided for the
problem associated with the appliance, wherein a provider of a
virtualization environment that was used to create the appliance
provides the support for the problem.
26. The system of claim 25, wherein the response includes
forwarding the support request to a partner associated with the
virtualization environment provider, wherein the partner provides
the support for the problem.
27. The system of claim 22, wherein the support engine is further
configured to deny the support request in response to determining
that support cannot be provided for the problem associated with the
appliance.
28. The system of claim 22, wherein the support engine is further
configured to: identify one or more changes that can be implemented
for the appliance to render the problem associated with the
appliance supportable; generate a response to the support request
that includes instructions for implementing the identified
changes.
29. The system of claim 22, further comprising a virtualization
environment configured to coordinate billing for the support
request.
30. The system of claim 22, wherein the support engine is further
configured to update a knowledge based that the support engine uses
to analyze supportability and service support requests in response
to resolving the support request.
31. A method for supporting a virtual appliance, comprising:
invoking a support engine in response to a build engine receiving a
request to build an image corresponding to an appliance;
identifying one or more software components installed for the
appliance using the support engine; analyzing the installed
software components in view of one or more rules that express
supportability guidelines for the installed software components;
and generating a support statement indicating whether or not
support can be provided for the installed software components.
32. The method of claim 30, wherein the one or more rules include
heuristics that describe one or more of permitted changes,
permitted modifications, or permitted variants for the installed
software components.
33. The method of claim 30, further comprising: retrieving metadata
describing the installed software components; and analyzing the
metadata describing the installed software components in view of
the one or more rules, wherein the one or more rules express the
supportability guidelines in terms of the metadata describing the
installed software components.
34. The method of claim 30, further comprising: receiving a support
request at the support engine, wherein support request describes a
problem associated with the appliance; and referencing the support
statement to determine whether or not support can be provided for
the problem associated with the appliance.
35. The method of claim 33, wherein the support request includes a
snapshot of the appliance in a state associated with the
problem.
36. The method of claim 34, further comprising generating a second
support statement indicating whether or not support can be provided
for the appliance in the problem state.
37. The method of claim 33, further comprising generating a
response to the support request in response to determining that
support can be provided for the problem associated with the
appliance, wherein a provider of a virtualization environment that
was used to create the appliance provides the support for the
problem.
38. The method of claim 36, wherein the response includes
forwarding the support request to a partner associated with the
virtualization environment provider, wherein the partner provides
the support for the problem.
39. The method of claim 33, further comprising denying the support
request in response to determining that support cannot be provided
for the problem associated with the appliance.
40. The method of claim 33, further comprising: identifying one or
more changes that can be implemented for the appliance to render
the problem associated with the appliance supportable; generating a
response to the support request that includes instructions for
implementing the identified changes.
41. The method of claim 33, further comprising coordinating billing
for the support request.
42. The method of claim 33, further comprising updating a knowledge
based that the support engine uses to analyze supportability and
service support requests in response to resolving the support
request.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional
Patent Application Ser. No. 61/057,593, entitled "System and Method
for Managing a Virtual Appliance Lifecycle," filed May 30, 2008,
the disclosure of which is hereby incorporated by reference in its
entirety.
FIELD OF THE INVENTION
[0002] invention relates to a system and method for supporting a
virtual appliance, and in particular, to an update server that
manages a rebuild and an integration test for an appliance in
response to detecting upstream updates to one or more software
components that have been installed for the appliance, and further
to a support analysis manager that analyzes the software components
that have been installed for the appliance in view of various
heuristic rules to generate a support statement indicating whether
support is available for the appliance.
BACKGROUND OF THE INVENTION
[0003] Virtualization generally refers to the concept of running
one or more virtual machines that can share resources of a physical
machine across various different environments. For example, in a
virtualized system, different virtual machines can execute
different operating systems and/or applications on the same
physical machine, regardless of underlying hardware associated with
the physical machine. One advantage that virtualization can provide
is the ability to create a tuned virtual appliance that includes a
fully configured application image with just enough operating
system (JeOS) components and/or other components needed to run a
particular application. For example, an independent software vendor
(ISV) may develop virtual appliances to bundle one or more
applications with an operating system into a virtual machine image
that can be delivered to a customer, wherein the virtual machine
image may include all of the information necessary to install and
configure the application. Virtualization may therefore reduce or
eliminate barriers associated with different computing
environments, providing ISVs with increased flexibility for
reaching new markets and customers.
[0004] As such, the concept of virtualization has important
implications in simplifying the production and distribution of
software by reducing or eliminating the need for hardware
optimization. For example, an operating system vendor may simply
create a virtual appliance that includes the components of the
operating system (with or without other applications), where
customers may then run the operating system within a hypervisor
that emulates the relevant hardware environment. Furthermore,
virtual appliances may be deployed in many different ways,
including in hosted environments (e.g., to deliver an application
as a service), cloud computing environments (e.g., to reduce the
resources needed to run an application), or various other
environments. Thus, virtualization provides important advantages
for ISVs that may lack particular expertise for different computing
environments, as virtual appliances may provide contained systems
that have been thoroughly tested and configured for a particular
application. Virtual appliances may also simplify issues relating
to support, security, and complexity of software products by
reducing the volume of operating system components, services, and
applications necessary to deliver an application to customers that
may have different computational requirements.
[0005] Despite the various advantages that virtualization can
provide to software providers, existing virtualization systems tend
to lack a simple yet repeatable process for creating virtual
appliances. For example, virtualization has faced various barriers
to adoption, including the perception that specific skills are
needed to develop and support virtual appliances. That is, to
create a virtual appliance, developers often have to possess
knowledge regarding repositories where the relevant JeOS components
or other software components may be available, as well as issues
pertaining to dependencies or incompatibilities among software
components. Moreover, a historic problem with open source software
projects is that one developer may innovate or otherwise improve
upon certain aspects of a project, yet the changes that the
developer implements may not necessarily be made available to
upstream distributions of the project. Alternatively, the changes
may not become available upstream by the time other developers have
begun to work on the same features or problems. As such, existing
systems for developing virtual appliances tend to lack adequate
mechanisms for simplifying the management of origin repositories
and relationships between software components, and further tend to
lack adequate mechanisms for sharing the work and knowledge of
different virtual appliance developers.
SUMMARY OF THE INVENTION
[0006] According to one aspect of the invention, a system and
method for managing a virtual appliance lifecycle may address these
and other drawbacks of existing systems. In particular, the systems
and methods described herein may employ a hosted virtualization
environment to provide a unified framework for developing,
building, executing, deploying, and otherwise managing a virtual
appliance lifecycle. As will be described in further detail below,
integrating the functions of developing, building, executing,
deploying, and otherwise managing the lifecycle of virtual
appliances within the unified virtualization environment may
provide various synergies. For example, the synergies provided
through the integrated functions of the virtualization environment
may include enabling users to follow a simple and repeatable
process to build and configure a virtual appliance distribution,
monitor upstream changes and modifications to virtual appliances in
a collaborative and user-driven manner, and provide real-time
analysis and feedback during various phases of the virtual
appliance lifecycle.
[0007] According to one aspect of the invention, the virtualization
environment may be provided as a hosted service, which one or more
users may access to create tuned virtual appliances containing
"just enough operating system" (JeOS) components to run particular
applications or perform specific functions, and further to build
images of the appliances that can be executed in any suitable
runtime environment. In particular, the hosted virtualization
environment may include a development environment that provides
various features and options for developing virtual appliances. For
example, the development environment may provide various JeOS
templates that may be cloned to create a base appliance, thereby
providing a starting point for developing a virtual appliance
(e.g., the JeOS templates may generally include only the core
components, packages, patterns, or other information needed to
create a virtual appliance having a minimal operating system
footprint). In addition, the development environment may further
provide an appliance marketplace having various appliances
previously developed in the virtualization environment, wherein the
appliances in the appliance marketplace may be cloned in a similar
manner to create the base appliance. As such, the appliance
marketplace may provide a community where users can collaborate and
otherwise interact to manage the lifecycle of virtual appliances.
Alternatively, instead of cloning one of the JeOS templates or
appliances in the appliance marketplace, a user may create an
appliance from an imported configuration file providing an
appliance image description (e.g., a "config.xml" file created
using an image creation system, such as the openSUSE Kiwi Image
System).
[0008] According to one aspect of the invention, the development
environment may load an editing interface for a user to further
develop an appliance cloned from a JeOS template, an appliance
cloned from the appliance marketplace, or an appliance created from
a predefined configuration file. For example, the user may interact
with the editing interface to specify a name, narrative
description, base distribution, architecture, origin, or other
information for the appliance. As such, the editing interface may
further enable the user to select one or more packages, patterns,
files, or other software components to be included in and/or banned
from the appliance.
[0009] According to one aspect of the invention, the virtualization
environment may be associated with a repository metadata server
that caches information associated with various origin repositories
(e.g., packages, patterns, or other software contained in the
origin repositories, metadata describing the origin repositories
and the packages, patterns, or other software contained therein,
etc.). As such, the editing interface may enable the user to
interact with the repository metadata server to add, remove,
update, or otherwise query any suitable information cached from or
otherwise associated with the origin repositories, and the user may
further interact with the repository metadata server to manage
software for the virtual appliance (e.g., selecting packages,
patterns, or other software to be included in and/or banned from
the appliance). In addition, to assist the user in managing the
software for the virtual appliance, the virtualization environment
may provide various metrics derived from activity associated with
other users (e.g., a popularity metric may indicate the popularity
of particular packages or origin repositories among various users,
a trust metric may indicate whether certain software or origin
repositories are provided by a trusted source, etc.).
[0010] According to one aspect of the invention, in addition to
caching information associated with the origin repositories, the
repository metadata server may further discover dependency
information for the origin repositories (e.g., the repository
metadata server may build resolution graphs that describe
dependencies for the packages, patterns, or other software in the
origin repositories). In particular, the repository metadata server
may include an active daemon list identifying one or more active
resolution daemons associated with a dependency resolution service,
wherein the repository metadata server may push the resolution
graphs to the active resolution daemons in the active daemon list.
Thus, the user may interact with the repository metadata server in
various ways to manage software for a virtual appliance (e.g., to
obtain a real-time status of the origin repositories, browse,
search, or otherwise select software to be included in and/or
banned from the virtual appliance, etc.), while the dependency
resolution service may use the resolution graphs provided by the
repository metadata server to analyze various dependencies for the
software selected by the user.
[0011] According to one aspect of the invention, in response to a
package or other software being added, banned, or otherwise
selected for the appliance, the dependency resolution service may
invoke a resolution daemon to identify any dependent, recommended,
suggested, and/or conflicting software. For example, the resolution
daemon may be configured to scan the resolution graphs for
annotations or metadata representing explicit dependencies for the
selected software, and further to scan pre-installation scripts,
post-installation scripts, content, and/or documentation for
implicit dependencies associated with the selected software. Thus,
in response to a user adding to the appliance a particular software
component requiring other components to be installed, the required
components may be automatically added to the appliance, while any
conflicting components previously added to the appliance may be
removed to resolve the conflict. Furthermore, in response to the
added component having recommended and/or suggested dependencies,
the user may be notified of such dependencies to enable the user to
determine whether or not to include the recommended and/or
suggested dependencies, or the recommended and/or suggested
dependencies may be optionally added to the appliance automatically
with the user being provided with an option to override or
otherwise undo the automatic addition.
[0012] According to one aspect of the invention, in response to the
packages or other software being added, banned, or otherwise
selected for the appliance, and further in response to one or more
dependent packages being added to and/or removed from the
appliance, an impact analysis engine may analyze any resulting
changes to the appliance. In addition, the impact analysis engine
may provide the user with an appropriate notification of the
resulting changes, wherein the notification may optionally enable
the user to override or otherwise undo one or more of the changes.
For example, the notification provided by the impact analysis
engine may include a list of packages added to and/or deleted from
the appliance, a data impact on the appliance expressed in terms of
megabytes, gigabytes, or other data quantities, an error correction
mechanism that can be selected to resolve an error, inconsistency,
or other issue, a warning relating to an error, inconsistency, or
other issue that requires manual intervention to correct, an undo
mechanism to remove a package that was added to the appliance, an
override mechanism to add a package that was banned or deleted from
the appliance, and/or other information that may be relevant to the
resulting changes. As such, the impact analysis engine may provide
the user with various visual notifications to assist the user in
managing software for the appliance.
[0013] According to one aspect of the invention, upon the user
having selected the packages, patterns, or other software
components to be included in the appliance, the editing interface
may further enable the user to configure a build for the appliance.
In particular, the editing interface may provide various options to
configure settings for locales, networks, identity management, run
levels, applications, programs, services, storage, memory, and/or
any other appropriate settings for the appliance and/or the
software included in the appliance. In addition, the configuration
options provided to the user may optionally be further customized
based on metadata or other information relating to the packages or
other software included in the appliance (e.g., in response to a
particular package being included in the appliance, the customized
configuration options may enable the user to configure specific
services associated with that package). As such, the user may
configure the build for the appliance such that a resulting image
of the appliance and the software included therein is configured to
operate in accordance with the configuration options provided by
the user. Furthermore, the impact analysis engine may analyze the
configuration options provided by the user to identify any changes,
errors, and/or warnings associated with the configuration options
(e.g., identifying a particular package required to enable a
specific configuration option).
[0014] According to one aspect of the invention, upon the user
having configured the appliance and/or any software having
customized configuration options, the editing interface may further
enable the user to personalize the build for the appliance. In
particular, the editing interface may provide various options to
personalize the appliance. For example, the personalization options
may enable the user to add custom files or archives to the
appliance, and further to specify a target path or directory where
the custom files or archives are to be installed, identity
management settings to control access to the custom files or
archives, or various other settings for the custom files or
archives. Additionally, the personalization options may further
enable the user to establish graphics for logos, icons, boot
screens, login backgrounds, desktop backgrounds, or other visual
features for the appliance. As such, the user extend personalized
branding functionality to the appliance, replace pre-existing
graphics with custom graphics to "re-brand" the appliance, remove
one or more of the pre-existing graphics (e.g., to comply with
trademark rights), or otherwise personalize the visual features of
the appliance (e.g., logos, brands, or other marks to be used on
distributable media created for the appliance).
[0015] According to one aspect of the invention, the virtualization
environment may further provide one or more application program
interfaces to enable users, third-party developers, or other
entities to develop custom plug-ins or applications for the
virtualization environment. For example, the custom plug-ins or
applications may be designed to configure specific features
associated with software, functionality, services, or other aspects
of an appliance, provide customized appliance creation processes,
provide social networking features (e.g., managing the appliance
marketplace, relationships among users, etc.), or otherwise modify
any suitable functionality that may be provided in the
virtualization environment. Thus, the application program
interfaces may enable developers to add, extend, or otherwise
modify various features and functions of the virtualization
environment, thereby leveraging collective knowledge that may be
held within a development community and enabling the virtual
appliance lifecycle to be managed in a customized and personalized
manner (e.g., as described in further detail in co-pending U.S.
patent application Ser. No. ______, entitled "System and Method for
Managing a Virtual Appliance Lifecycle," filed on an even date
herewith, the disclosure of which is hereby incorporated by
reference in its entirety).
[0016] According to one aspect of the invention, the virtualization
environment may further include a build engine, which may be
configured to build images (e.g., operating system images) of
appliances that the user created, configured, personalized, and/or
otherwise developed within the development environment. In
particular, the build engine may generally build images that
provide fully configured systems ready to execute the software that
the user has installed for the appliance in response to being
deployed to a runtime environment.
[0017] According to one aspect of the invention, the build engine
may be invoked in response to receiving a build request from a
user, wherein the build request may generally include an image
description for the appliance (e.g., installed software, source
repositories for the software, configurations for the software
and/or the appliance, etc.). In addition, the build request may
further include an image format (e.g., a disk image, a live CD/DVD
image, a virtual machine image, etc.). The image description may be
validated to identify the appropriate information needed to build
the image (e.g., origin repositories containing the software,
directory structures, etc.). In response to the image description
containing any errors or other invalid information, the user may be
provided with a warning and/or an option to resolve the errors
(e.g., installing a package required for a specific software
configuration).
[0018] According to one aspect of the invention, the build engine
may be launched in response to the image description being
validated. In particular, launching the build engine may include
loading a virtual machine from a virtual machine pool to create a
contained environment for building the image, wherein the contained
environment may enable the build engine to build various images
corresponding to various different appliances in a substantially
simultaneous and independent manner. The build engine may then
analyze the image description and create a physical extend based on
the image description within the contained build environment (e.g.,
a file system layout having a directory structure and a file
layout). In response to successfully creating the physical extend
and the corresponding file system layout, the image (or logical
extend) may be created from the physical extend. For example,
creating the image may generally include retrieving the files to be
installed from an appropriate source repository and then copying
the files to the image in accordance with the file system layout.
As such, the image may comprise one file encapsulating all of the
software selected for the appliance installed according to the file
system layout, with the image being further configured and
personalized according to any parameters that the user may have
provided for the appliance.
[0019] According to one aspect of the invention, the build engine
may monitor the build for the image, whereby the build engine may
be further configured to generate one or more logs describing the
build process. In addition, the build engine may display the
current status for the build within a user interface (e.g., an
elapsed time, a progress bar, a current process being executed,
etc.). As such, in response to the build process completing
(whether successfully or unsuccessfully), one or more build logs
may be generated to provide a record describing the various
processes executed during the build, any errors, problems, or other
status changes that occurred during the build, any files or other
information referenced to create the image, or any or other
information describing the build process.
[0020] According to one aspect of the invention, an appliance
family tree may be constructed in response to the image having been
built for an appliance that was created from a cloned appliance
(e.g., a JeOS template, an appliance previously created by the
user, an appliance published in the appliance marketplace, etc.).
In particular, as noted above, the virtualization environment may
provide an appliance marketplace within a panel or other interface
that can be browsed and/or searched for various appliances that
users of the virtualization environment have made publicly
available. As such, various relationships between the cloned
appliance and the image derived from the cloned appliance may be
analyzed to construct the family tree. For example, the family tree
may represent links between the cloned appliance and any appliances
derived from the cloned appliance according to a hierarchy, and
changes or other differences among the linked appliances may be
represented along branches of the hierarchy, thereby enhancing the
information provided in the appliance marketplace. Thus, the
appliances linked to one another in the family tree may be analyzed
according to one or more data mining algorithms, heuristic
algorithms, and/or other analytic techniques to derive knowledge
relating to the changes or other differences among the appliances
(e.g., defining clusters of derived appliances that incorporate
similar changes to a cloned appliance, identifying "best changes,"
"recommended changes," and/or "popular changes," etc.).
[0021] According to one aspect of the invention, the virtualization
environment may further enable the user to manage publication of
the image. In particular, a publication prompt may be presented to
the user that created the appliance, wherein the publication prompt
may enable the user to establish whether or not the image of the
appliance is to be made publicly visible. Thus, in response to the
user indicating that the appliance is to be published, the image
may be added to the appliance marketplace, thereby enabling other
users to view and optionally clone the published appliance.
Alternatively, in response to the user not publishing the
appliance, the image will not be visible to other users of the
virtualization environment (although the unpublished image may be
made visible to one or more limited users, such as other users in a
friend list for the user, other users that created earlier
generations of the unpublished appliance, and/or other users that
created appliances in the same family tree as the unpublished
appliance). In either case, the image may be added to the user's
appliances, wherein the user may be provided with various options
to manage the user's appliances (e.g., viewing information
describing the appliances, deleting, downloading, cloning,
publishing, deploying, and/or otherwise managing any of the
appliances, etc.).
[0022] According to one aspect of the invention, the virtualization
environment may further include a hosted runtime environment for
executing virtual appliances. In particular, the hosted runtime
environment may be launched in response to a request that includes
an identifier for an image to be executed, wherein launching the
runtime environment may include loading a virtual machine from a
virtual machine pool to provide a contained runtime environment.
For example, one or more of the virtual machines in the virtual
machine pool may be in a pre-configured state corresponding to
snapshots captured during an initialized state for the runtime
environment to reduce an amount of time needed to initiate
executing the image, while providing the contained runtime
environment may permit substantially simultaneous and independent
execution for various different images corresponding to various
different appliances.
[0023] According to one aspect of the invention, a "snapshot" of
the image being executed in the hosted runtime environment may be
captured at any point during the execution. The snapshot may
generally capture an entire state of the image at any particular
point, wherein the snapshot may then be used in various ways to
manage the appliance. For example, the snapshot may be captured
after various tasks have been performed, such that a state of the
image can be captured at any particular time to enable restoring
the appliance to a runtime state corresponding to the snapshot. In
another example, the snapshot may be captured when the appliance is
in a crashed state to simplify debugging or support processes for
the appliance. Furthermore, various snapshots for the image may be
captured at different points in time, wherein any changes or other
differences between different snapshots may be attributed to
runtime activity that occurred in a period of time between when the
snapshots were captured. In this manner, different snapshots taken
at different points in time may be compared to one another to
provide feedback for auditing one or more specific activities.
[0024] According to one aspect of the invention, the hosted runtime
environment may be further configured to generate one or more
runtime logs. For example, the hosted runtime environment may
include a monitoring engine, which may be used to generate runtime
logs identifying files that were modified, unused, and/or underused
during execution of the appliance image. In particular, in one
implementation, the runtime logs may include any files that were
modified subsequent to the image being loaded in the runtime
environment, and may further include information specifically
describing the manner in which such files were modified.
Furthermore, the runtime logs may include any files that were
unused or underused subsequent to the image being loaded in the
runtime environment, and may further include information
specifically describing the manner in which such files were unused
or underused. As such, the user may interact with the runtime logs
and determine whether or not to modify the appliance. For example,
building a modified version of the appliance may result in an image
encapsulating one or more of the snapshots captured for the
appliance. In addition, the user may individually select one or
more changes in the runtime logs to apply in the modified version
of the appliance.
[0025] According to one aspect of the invention, the virtualization
environment may further include a support engine configured to
manage updates and supportability for managed appliances. In
particular, as noted above, the repository metadata server may
detect one or more upstream updates in response to local digests
for one or more origin repositories not matching current digests
for the one or more origin repositories. The support engine may
then invoke an update server to identify specific packages,
patterns, or other software components associated with the upstream
update and identify one or more of the managed appliances that
include the updated software components. The update server may then
generate a workflow to update the identified appliances, wherein
the workflow may generate a notification indicating that new
versions have been released for one or more software components for
the appliance. The workflow may further include rebuilding images
for the appliances to install the updated software components and
managing an integration test to ensure that the rebuilt images
function or otherwise behave correctly with the updated software
components installed. In response to determining that the rebuilt
appliance functions correctly, the rebuilt appliance may be
released to subsequent users that clone or otherwise deploy the
rebuilt appliance. In addition, the update server may generate one
or more updates that can released to earlier versions of the
rebuilt appliance, wherein the updates may be configured to install
the updated software components for the earlier versions of the
appliance.
[0026] According to one aspect of the invention, the support engine
may further include a support analysis manager configured to
analyze supportability and service support requests for one or more
of the appliances managed in the virtualization environment. In
particular, the support analysis manager may generally be
configured to analyze the managed appliances and generate
corresponding support statements indicating whether a provider of
the virtualization environment can support certain components that
have been installed for the appliances and/or certain problems that
may be encountered for the appliances. For example, if the support
analysis manager determines that the appliance substantially alters
otherwise supported software components distributed by the
virtualization environment provider, the support statement may
indicate that the virtualization environment provider cannot
support problems relating to the modified components. However, if
the appliance includes additional or different versions of certain
packages without substantially altering the functionality or
behavior of the supported components, the support statement may
indicate that problems relating to the supported components can be
supported, while support cannot be provided for any altered
components or components distributed from unknown or unsupported
sources.
[0027] According to one aspect of the invention, the support
analysis manager may further determine whether one or more trusted
partners or other entities known to the virtualization environment
can provide support for the components installed for the appliances
and/or the problems encountered for the appliances. Thus, the
support statement generated by the support analysis manager may
further indicate whether any relevant trusted partners or other
known entities that can provide the appropriate support. For
example, a cryptographic key may be used to indicate whether a
particular package, pattern, or other software component originates
from a "trusted" source, wherein the cryptographic key for the
trusted package may be configured to break or otherwise cease to be
intact in response to substantial alterations. Thus, the support
analysis manager may determine whether the cryptographic key or
other relevant data remains intact, wherein the determination may
be considered in combination with one or more heuristic rules to
generate the appropriate support statement.
[0028] According to one aspect of the invention, the support
analysis manager may be integrated with the build engine, wherein
the build engine may invoke the support analysis manager to analyze
supportability for an appliance currently being built. For example,
the virtualization environment may provide one or more guidelines
regarding supportability for various supported packages, patterns,
and other software components to describe certain changes or
modifications that can or cannot be implemented for the supported
components. As such, during a build for a particular appliance, the
support analysis manager may analyze the appliance to ensure that
the appliance adheres to the supportability guidelines and generate
the appropriate support statement. For example, each of the
components installed for the appliance may be analyzed in view of
various heuristic rules to validate whether the appliance adheres
to the supportability guidelines, wherein the heuristic rules may
express the supportability guidelines in terms of metadata to
specify permitted variants for the supported components. As such,
the support analysis manager may retrieve metadata for each of the
installed components in view of various heuristic rules to generate
the support statement indicating whether or not the supportable
software components have been modified to the extent of being
rendered unsupportable.
[0029] According to one aspect of the invention, the support
analysis manager may be further configured to service support
requests for one or more appliances managed in the virtualization
environment. In particular, users may submit one or more support
requests to the virtualization environment in response to
encountering problems with custom appliances, wherein the support
requests may include a snapshot of the appliance for which support
is requested and a description of the problem with the appliance.
As such, the support analysis manager may analyze the snapshot,
which may encapsulate an entire environment associated with the
problem state of the appliance, and then determine whether or not
the particular problem can be supported. For example, the support
analysis manager may reference an original support statement
generated when the appliance was built in the virtualization
environment, and may further analyze the appliance snapshot to
generate a current support statement for the problem state
associated with the appliance. The support analysis manager may
then determine whether the provider of the virtual environment
supports the problem described in the support request and generate
a response to the support request in response to determining that
the virtualization environment provider supports the problem.
[0030] According to one aspect of the invention, in response to
determining that the virtualization environment provider cannot
support the problem identified in the support request, the support
analysis manager may further determine whether upstream support can
be provided for the problem by one or more trusted partners or
other known entities. In response to the original support statement
and/or the current support statement indicating that the problem
can be supported by the trusted partners and/or other known
entities, the support request may be routed to the appropriate
supporting entity, thereby coordinating support to resolve the
problem. Alternatively, the support request may be denied in
response to determining that the virtualization environment
provider, the trusted partners, and the other known entities cannot
support the problem (e.g., because the problem relates to a custom
component that the user has installed, a component distributed by a
third-party not associated with the virtualization environment, a
supported component that has been substantially altered, etc.).
Furthermore, in response to denying the support request, the
support analysis manager may further determine whether a change or
other modification can be implemented for the relevant component
causing the problem in a manner that renders the problem
supportable. For example, if the problem is unsupportable because a
particular package installed for the appliance alters functionality
or behavior of the operating system, the denial of the support
request may include instructions indicating that the problem may be
supportable in response to uninstalling the package.
[0031] According to one aspect of the invention, in response to
appropriately servicing the support request (i.e., either
responding to the support request, routing the support request, or
denying the support request), feedback may then be provided to the
support engine. For example, the feedback may include information
relating to whether the virtualization environment provider, the
trusted partners, and/or the other known entities successfully or
unsuccessfully resolved a supported problem, wherein such
information may be referenced to resolve similar problems that
arise subsequently, modify the heuristic rules or supportability
guidelines, or otherwise add to a knowledge base that the support
engine employs to analyze supportability and/or service support
requests. In another example, the feedback may include information
relating to problems for which support was denied, wherein such
information may be further referenced to determine whether or not
the heuristic rules or supportability guidelines should be
modified. For example, if a substantial number of users experience
similar problems with a particular unsupported package, the
virtualization environment provider may determine that support for
the package should be added, whereby one or more appropriate
entities may be enlisted to provide support for the package, and
further to create appropriate heuristic rules and supportability
guidelines for the package.
[0032] According to one aspect of the invention, the virtualization
environment may further enable appliances to be deployed to one or
more runtime environments. For example, a user may download one or
more images to a client device for execution in an appropriate
runtime environment at the client device. In addition, one or more
images may be deployed to a web server, an Amazon Elastic Compute
Cloud (EC2), or any other environment suitable for hosting or
otherwise executing the appliance. Alternatively (or additionally),
the user may execute an appliance in the hosted runtime
environment, in which case the hosted runtime environment may be
launched and managed in a similar manner as discussed above.
[0033] Other objects and advantages of the invention will be
apparent to those skilled in the art based on the following
drawings and detailed description.
BRIEF DESCRIPTION OF THE DRAWINGS
[0034] FIG. 1 illustrates a block diagram of an exemplary system
for managing a virtual appliance lifecycle, according to one aspect
of the invention.
[0035] FIG. 2 illustrates a flow diagram of an exemplary method for
creating virtual appliances, according to one aspect of the
invention.
[0036] FIG. 3 illustrates a flow diagram of an exemplary method for
building virtual appliances, according to one aspect of the
invention.
[0037] FIG. 4 illustrates a flow diagram of an exemplary method for
testing virtual appliances in a runtime environment, according to
one aspect of the invention.
[0038] FIG. 5 illustrates a flow diagram of an exemplary method for
managing updates for a virtual appliance, according to one aspect
of the invention.
[0039] FIG. 6 illustrates a flow diagram of an exemplary method for
managing support for a virtual appliance, according to one aspect
of the invention.
DETAILED DESCRIPTION
[0040] According to one aspect of the invention, FIG. 1 illustrates
an exemplary system for managing a virtual appliance lifecycle. In
particular, the system illustrated in FIG. 1 may include a
virtualization environment 120, which provides a unified framework
for developing, building, executing, deploying, and otherwise
managing a lifecycle for virtual appliances. Integrating the
functions of developing, building, executing, deploying, and
otherwise managing the lifecycle of virtual appliances within the
unified virtualization environment 120 may provide various
synergies, including the ability to enable users to easily follow a
repeatable process to create, build, and configure a virtual
appliance distribution, monitor upstream changes and modifications
to virtual appliances in a collaborative and user-driven manner,
and provide real-time analysis and feedback during various phases
of the virtual appliance lifecycle. In one implementation, for
example, the virtualization environment 120 may be used to create
tuned virtual appliances that contain one or more applications and
"just enough operating system" (JeOS) components to run the one or
more applications, create live portable media (e.g., a CD, DVD, USB
key, etc.) that contain particular packages or software, and/or
build hard disk images to be preloaded directly onto hardware,
among various other tasks relating to management of a virtual
appliance lifecycle, as described in further detail below.
[0041] In one implementation, the virtualization environment 120
illustrated in FIG. 1 may be provided as a hosted service, wherein
a local or remote client device 105 may access the virtualization
environment 120 over a network 110a to develop one or more virtual
appliances. The client device 105 may then access a development
environment 130 associated with the hosted virtualization
environment 120, wherein the development environment 130 may be
configured to operate in a create mode that enables a user of the
client device 105 to develop one or more virtual appliances. In
particular, when the user initiates the create mode, the
development environment 130 may be configured to launch a
development interface within a user interface 125, wherein the
development interface may provide various options for the user to
develop one or more virtual appliances.
[0042] For example, in one implementation, the development
interface may provide the user with an option to create a virtual
appliance from one or more "just enough operating system" (JeOS)
templates, which may be stored in an appliance repository 180
coupled to the virtualization environment. In particular, each of
the JeOS templates may generally include only the core components,
packages, patterns, or other information needed to create a virtual
appliance having a minimal operating system footprint. In one
implementation, in response to the user selecting the option to
create an appliance from the JeOS templates, the JeOS templates
available in the appliance repository 180 may be presented to the
user for selection within the user interface 125. For example, the
user may select a text only JeOS template to create an appliance
having nothing more than a general purpose Linux operating system
(i.e., with no graphical interface). Alternatively (or
additionally), the user may select a JeOS template that includes a
general purpose operating system and one or more additional
components (e.g., one or more of the JeOS templates may provide a
minimal graphical interface, a GNOME or KDE desktop environment, a
virtual server package, or other components in addition to the
general purpose Linux operating system). As such, in response to
the user selecting one of the JeOS templates, the development
environment 130 may create a new base appliance including only the
components that correspond to the selected template, and the new
base appliance may then be loaded for further development within
the user interface 125.
[0043] In one implementation, the development interface may further
provide the user with an option to clone an appliance that the user
and/or other users have previously developed in the virtualization
environment 120. For example, in one implementation, the appliance
repository 180 may store one or more appliances that the user
previously developed in the virtualization environment 120, in
addition to various appliances that users of the virtualization
environment 120 have published for public access. As such, the
development environment 130 may provide the user with an option to
browse and/or search an appliance marketplace, which may include
the user's previously developed appliances and the published
appliances stored in the appliance repository 180.
[0044] For example, in one implementation, the appliance repository
180 may include a descriptive information for each of the
appliances available in the appliance marketplace, including
appliance names, developer names or aliases, descriptions of
software or other components in the appliances, release information
(e.g., version numbers, numbers of releases, release dates, release
types, release sizes, etc.), popularity (e.g., numbers of positive
and/or negative user votes, numbers of user downloads, user
comments, etc.), or other information describing the various
appliances in the marketplace. Thus, the user may browse the
appliance marketplace to select an appliance from the appliance
repository 180 or search the appliance marketplace to identify
appliances having certain characteristics or matching certain
criteria. For example, the user may search the appliance
marketplace according to appliance name, software, release
information (e.g., age, release type, version number, etc.), size
(e.g., a number of megabytes or gigabytes), and/or popularity
(e.g., voting statistics, download statistics, etc.). In another
example, the user may create a profile describing a particular
hardware configuration and then search for appliances in the
marketplace that are compatible with that hardware configuration,
popular among other users having similar hardware configurations,
or otherwise relevant to the information in the profile.
[0045] In one implementation, in response to the user selecting one
of the appliances in the marketplace after browsing and/or
searching the appliance marketplace, the development environment
130 may clone the selected appliance to create the base appliance
that can be further developed within the user interface 125.
Alternatively (or additionally), the user may download the
appliances in the marketplace to the client device 105, provide
positive or negative votes for the appliances in the marketplace,
and/or post comments in a user-driven discussion for the appliances
in the appliance marketplace. As such, in addition to providing the
user with access to other appliances that have been developed in
the virtualization environment 120, the appliance marketplace may
provide a community in which users can collaborate and otherwise
interact to manage the lifecycle of a virtual appliance.
[0046] In one implementation, the development interface may further
provide the user with an option to import a predefined
configuration file, wherein the development environment 130 may be
configured to create the base appliance from the imported
configuration file. For example, in one implementation, the
development environment 130 may be capable of importing and parsing
configuration files associated with any suitable image creation
system, wherein the configuration files generally include an image
description for an appliance (i.e., name, author, contact
information, description, and version), a definition of users,
groups, or other identity management settings for the appliance, a
definition of repositories and packages included in the appliance,
and a directory structure for the appliance image. Thus, in
response to the user importing a configuration file, the
development environment 130 may parse the configuration file,
create the base appliance from the configuration file, and then
load the base appliance for further development within the user
interface 125. For example, in one implementation, the development
environment 130 may be configured to import configuration files
defined in accordance with an image creation system, such as the
openSUSE Kiwi Image System, as described in "openSUSE-KIWI Image
System Cookbook" by Marcus Schaefer, the contents of which is
hereby incorporated by reference in its entirety.
[0047] In one implementation, upon loading the newly created base
appliance for further development within the user interface 125, an
appliance overview describing the appliance may be displayed,
wherein the user can edit one or more aspects of the description.
For example, if the user created the new appliance from the text
only JeOS template described above, the description of the
appliance may include a name (e.g., "My Custom Appliance"), a brief
narrative description (e.g., "A basic openSUSE 10.3 system, with no
graphical interface"), a base distribution (e.g., openSUSE 10.3),
an architecture (e.g., x86 and x86.sub.--64), an origin (e.g.,
openSUSE 10.3 and the text only JeOS template), and various
statistics (e.g., a number and type for various stored builds of
the appliance, descriptions for any of the stored builds that have
been published in the marketplace, etc.). Thus, the manner in which
the appliance was created and/or cloned may be used to populate the
descriptive information such as base distribution, architecture,
and origin, while the user may edit the appliance name, the
narrative description, or any other information that does not
directly result from the manner in which the appliance was created
and/or cloned.
[0048] In one implementation, once the appliance has been loaded
for further development and the user has completed the appliance
overview, the create mode may enable the user to select various
packages, patterns, files, or other software components to be
included in the appliance. For example, packages may generally
include distributions of related software components and
descriptive metadata such as a full name, version number, vendor,
and dependencies for the software, and patterns may generally
include reusable descriptions or templates for solutions that
address recurring software design problems. Thus, in one
implementation, a repository metadata server 190 may be configured
to cache information associated with one or more origin
repositories 115 that include packages, patterns, or other software
that can be added to appliances developed in the virtualization
environment. Further detail describing the repository metadata
server 190 is included in co-pending U.S. patent application Ser.
No. 12/369,188, entitled "System and Method for Building Virtual
Appliances Using a Repository Metadata Server and a Dependency
Resolution Service," filed Feb. 11, 2009, the disclosure of which
is hereby incorporated by reference in its entirety.
[0049] In particular, the repository metadata server 190 may be
configured to contact one or more of the origin repositories 115
over a network 110b. The repository metadata server 190 may then
cache information contained in the origin repositories 115, wherein
the cached information may include the packages, patterns, or other
software contained in the origin repositories 115, metadata
describing the origin repositories 115, and metadata describing the
packages, patterns, or other software contained in the origin
repositories 115. As such, the user may interact with the
repository metadata server 190 to add, remove, update, or otherwise
query information associated with one or more of the origin
repositories 115. For example, the repository metadata server 190
may communicate with the client device 105 and/or the origin
repositories 115 using function calls associated with XML-RPC,
SOAP, local UNIX pipes, a custom communication implementation, or
other communication mechanisms to enable the user to select
packages, patterns, or other software components contained in the
origin repositories 115.
[0050] For example, in one implementation, the create mode may
provide the user with an option to add a new origin repository 115,
wherein the user may define a name for the new origin repository
115 and further provide a location where information for the new
origin repository 115 can be obtained. The repository metadata
server 170 may then contact the new origin repository 115 using the
location provided by the user and retrieve metadata describing the
new origin repository 115 and any packages and patterns contained
therein. In one implementation, the metadata retrieved from the
origin repositories 115 may be used to build resolution data files
that can be provided to a dependency resolution service 140. For
example, as described in further detail in co-pending U.S. patent
application Ser. No. 12/369,188 referenced above, the repository
metadata server 190 may include an active daemon list identifying
one or more active resolution daemons associated with the
dependency resolution service 140. Thus, the repository metadata
server 190 may build the resolution data files, which generally
include one or more resolution graphs, and the resolution data
files may be pushed to the active resolution daemons in the active
daemon list.
[0051] In one implementation, the repository metadata server 190
may further insert the metadata describing the new origin
repository 115 and the packages and patterns contained therein into
a repository database 145 coupled to the dependency resolution
service 140. As such, the development environment 130 may reference
information relating to the origin repository 115 through the
repository database 145, such that the user can access the origin
repository 115 and the packages, patterns, or other software
contained therein through the development environment 130.
Additionally, the repository metadata server 190 may download the
packages, patterns, or other software from the origin repository
115 and then store the download packages, patterns, or other
software in one or more cached repositories 195, thereby creating a
cached version of the origin repository 115 stored locally with
respect to the virtualization environment 120.
[0052] In one implementation, the create mode may further provide
the user with an option to select one or more of the origin
repositories 115 available through the repository metadata server
190. The selected repositories 115 may then provide a source of
software that can be included in the appliance. For example, the
user may query the repository database 145 to view information
relating to the origin repositories 115 available through the
repository metadata server 190 (including any repositories 115 that
may have been added by the user), or the user may search the
repository database 145 to identify one or more of the origin
repositories 115 having a particular name, a particular type, or
particular software (e.g., the user may enter a search string of
"Apache" to search the repository database 145 for origin
repositories 115 that contain one or more Apache packages).
Furthermore, in response to determining that a plurality of the
origin repositories 115 contain software matching the search
criteria provided by the user, the plurality of origin repositories
115 may be ranked in various ways (e.g., based on a number of
packages matching the search within each of the various origin
repositories 115, popularity of the various origin repositories
115, etc.). In response to the user selecting one or more of the
origin repositories 115 to be a source of software for the virtual
appliance, the repository metadata server 190 may then enable the
user to access the packages and patterns contained in the selected
origin repositories 115 (e.g., by downloading the packages and
patterns to the cached repositories 195, and enabling the user to
retrieve the downloaded packages and patterns from the cached
repositories 195).
[0053] In one implementation, to provide reliable responses to user
queries, the repository metadata server 190 may be configured to
ensure that the cached repositories 195 maintain an updated cache
of the origin repositories 115. For example, if a user requests
information relating to a particular one of the origin repositories
115, the repository metadata server 190 may compare a local digest
for the origin repository 115 to a current digest for the origin
repository 115 (e.g., the digest may represent a metadata signature
for the origin repository 115, wherein any changes to the origin
repository 115 may result in a change to the digest or metadata
signature for the origin repository 115). Alternatively (or
additionally), the repository metadata server 190 may compare the
local digests for one or more of the origin repositories 115 to
current digests for the origin repositories 115 according to a
schedule, which may be predetermined, user-defined, or configured
in various other ways. As such, if the local digest for the origin
repository 115 matches the current digest for the origin repository
115, the repository metadata server 190 may determine that the
cached repositories 195 are current for the origin repository 115.
However, if the local digest and the current digest do not match,
the repository metadata server 190 may follow the procedures
described above for adding a new origin repository 115 to update
the outdated cache of the origin repository 115.
[0054] In one implementation, the user may interact with the
repository metadata server 190 in various ways to manage software
for a virtual appliance. For example, the repository metadata
server 190 may be queried to obtain a real-time status of the
origin repositories 115, including whether a particular origin
repository 115 is currently being added, updated, or deleted. The
user may also browse the origin repositories 115 in various ways to
select software to be included in the virtual appliance. For
example, software in the origin repositories 115 may be organized
into various software groups, wherein the user may browse the
groups to select packages, patterns, or other software components
that provide specific functionality (e.g., the groups may organize
software based on functionality relating to operating systems,
development, games, graphics, multimedia, office suites, systems,
servers, networking, desktop environments, etc.). In one
implementation, the user may further interact with the repository
metadata server 190 to search the origin repositories 115 for
packages, patterns, or other software components that match certain
criteria provided by the user.
[0055] As such, the repository database 145 may enable the user to
browse and/or search the software contained in the origin
repositories 115 in order to manage software for a virtual
appliance (e.g., the user may select packages, patterns, or other
software to be included in the appliance, and may further ban
packages, patterns, or other software from the appliance). For
example, in addition to populating the repository database using
the metadata associated with the packages, patterns, or other
software downloaded from the origin repositories, the repository
metadata server may further scan the downloaded software to
discover additional information describing the software. As such,
the repository database 145 may further respond to queries from the
user relating to the information discovered within the downloaded
software (e.g., the metadata may indicate that a particular package
has components for configuring users for a particular system, such
as an Apache or PostgreSQL service). In one implementation,
software within the repository database 145 may be filtered
according to one or more active origin repositories 115 selected by
the user, subsets or groups of the active origin repositories 115,
and/or all of the origin repositories 115 known to the repository
metadata server 190. Additionally, information relating to the
origin repositories 115 selected by the user may be merged with the
appliance to provide dynamic control over updating the appliance
(e.g., the appliance may be automatically updated in response to an
update to an active origin repository 115 or an update to a
relevant package contained in the active origin repository 115, or
a message may be sent to the user that such an update is available,
etc.).
[0056] In one implementation, to assist the user in making
selections relating to the packages, patterns, or other software
available in the origin repositories 115, the software may be
associated with one or more metrics based on activity in the
virtualization environment 120. For example, a user may search the
repository database 145 for a particular package, and packages
matching the search may be ranked based on a popularity metric for
the matching packages. In one implementation, the popularity metric
may be derived from how popular a particular package or origin
repository 115 is among various users of the virtualization
environment 120, thereby providing information to distinguish
different packages or versions of a package based on selections
made by other users. For example, the popularity metric may be
provided as a numeric integer such as a ranking from one to ten, a
visual object such as a series of vertical lines (e.g., with the
number of vertical lines increasing as a function of popularity),
or in various other ways.
[0057] In one implementation, the metrics may further include a
trust metric for one or more of the origin repositories 115 and/or
software contained therein, wherein the trust metric may provide
the user with information identifying trusted software or origin
repositories 115. In particular, the repository metadata server 190
may provide an interface for an administrator or another authorized
user to manage cryptographic keys (e.g., PGP certificates) that are
used to determine whether a particular package or origin repository
115 is "trusted" or "untrusted." For example, if the repository
metadata server 190 downloads a package signed with a cryptographic
key from a particular origin repository 115, the cryptographic key
may be validated to determine the trust metric for the particular
package or origin repository 115 (e.g., if the cryptographic key
indicates that the package originates from a trusted partner or
other entity known to a provider of the virtualization environment
120, the package or repository 115 may be marked "trusted," whereas
if the cryptographic key indicates that the package originates from
an unknown source, which may or may not be malicious, the package
or repository 115 may be marked "untrusted").
[0058] In one implementation, the user interface 125 for searching
the origin repositories 115 may include an automatic complete
feature to identify potentially relevant search results. For
example, when the user begins typing a search string within the
search user interface 125, the metadata stored in the repository
database 145 may be queried in real-time to identify the
potentially relevant results, prior to the user completing and
submitting the search string. As such, the potentially relevant
results may then be used to automatically complete the search
string. For instance, if the user wants to add an IMAP server
package to the appliance but does not know the correct name of the
package, the user may begin typing "IMAP . . . " within the search
user interface 125. The repository metadata database 145 may then
be searched to determine one or more search strings that begin with
or otherwise contain "IMAP." The search strings may then be
displayed to recommend one or more completion terms (or
combinations of completion terms), wherein the user can select one
of the recommended completion terms (or combinations of completion
terms) to automatically complete the search string that the user is
currently typing. As such, the automatic complete feature may
provide a mechanism to search for software in a manner that
tolerates minor misspellings, variations in terminology, and/or
incomplete information, while also employing search engine
capabilities to recommend one or more terms to complete the search
string being provided by the user.
[0059] In one implementation, the user interface 125 for
interacting with the software in the origin repositories 115 may be
configured to provide information describing the software in
response to one or more interactions with the client device 105.
For example, the user at the client device 105 may hold a mouse
pointer or other input mechanism over a certain package or software
component, and a pop-up window may be displayed in the user
interface 125 in response to the interaction. In one
implementation, the pop-up window may include information such as
an origin repository 115, a version number, one or more groups
(e.g., productivity, networking, browsers, etc.), a size or other
data quantity, software licenses and permissions, or other
information describing the package or software component.
Alternatively (or additionally), the user may click or otherwise
select the package or software component to view a more detailed
description of the package or software component.
[0060] In one implementation, in response to the user selecting one
or more packages, patterns, or other components for the appliance,
the selected components may be retrieved from the cached
repositories 195 and added to the appliance. In one implementation,
if the repository metadata server 190 has not yet downloaded the
selected components to the cached repositories 195, the selected
components may be downloaded to the cached repositories 195 and
then added to the appliance. Further, the repository metadata
server 190 may dynamically reorder a download queue to prioritize
the selected components. Additionally, the user may query the
repository metadata server 190 to obtain a current status of any
pending downloads (e.g., the current status may identify a package
or other component currently being downloaded, a number of packages
or other components remaining to be downloaded, or other
information relating to the status of software being downloaded
from the origin repositories 115 to the cached repositories
195).
[0061] In one implementation, in response to the user selecting
and/or banning software for the appliance, the dependency
resolution service 140 may scan the resolution graphs provided by
the repository metadata server 190 to analyze various dependencies
for the selected and/or banned software. For example, any
particular package may include metadata describing, among other
things, other packages that are required for the package to operate
properly, other packages that are recommended (but not required)
for the package to operate optimally, other packages that are
suggested (but not required) to enhance the package's
functionality, and other packages that conflict (and cannot be
installed) with the package. As such, if the user selects a
particular package to be included in the appliance, the dependency
resolution service 140 may scan the metadata associated with the
selected package to obtain a complete manifest of dependent
packages. The user may then be notified in response to other
packages being required for the selected package, other packages
already selected for the appliance conflicting with the selected
package, or other relevant dependencies being identified.
[0062] In one implementation, the repository metadata server 190
may and the dependency resolution service 140 may interact with one
another to manage software for appliances created in the
development environment 130. For example, the repository metadata
server 190 may register one or more active resolution daemons
associated with the dependency resolution service 140 and store
information relating to the registered resolution daemons in an
active daemon list. As such, when the repository metadata server
190 adds, updates, or deletes information associated with one or
more of the origin repositories 115, the repository metadata server
190 may push the current information associated with the origin
repositories 195 to the dependency resolution service 140, thereby
ensuring that the repository database 145 contains current
information for the origin repositories 115. In addition, the
repository metadata server 190 may use metadata retrieved from the
origin repositories 115 to create one or more resolution graphs
defining dependencies for the packages contained in the origin
repositories 115. The repository metadata server 190 may then push
the resolution graphs to the registered resolution daemons to be
used when resolving dependencies for software that the user has
selected and/or banned from the appliance.
[0063] In one implementation, in response to the user selecting
and/or banning one or more packages, patterns, or other software
components for the appliance, the dependency resolution service 140
may invoke one or more of the resolution daemons to identify
software that is dependent, recommended, suggested, and/or
conflicting for the selected and/or banned software. For example,
to resolve dependencies for a particular package, the active
resolution daemons may scan the resolution graphs for annotations
or metadata explicitly describing dependencies for the package. In
addition, the active resolution daemons may further scan
pre-installation and post-installation scripts for the package to
determine configurations for the scripts (e.g., a script may be
configured to execute a command relating to a dependent package,
and may further scan content associated with the package to
introspectively identify dependencies (e.g., based on documentation
included in the package).
[0064] Thus, when a user adds a particular package to the
appliance, the dependency resolution service 140 may identify any
other packages or components that the package may require, wherein
the required packages or components may be added to the appliance
automatically (e.g., in response to the user adding a server
package, one or more packages needed to authenticate users, manage
a file system, or perform other tasks to properly operate the
server may be added to the appliance automatically). Further, the
dependency resolution service 140 may identify any packages already
added to the appliance that conflict with the added package,
wherein the conflicting packages may be removed from the appliance
to resolve the conflict. In one implementation, if the added
package has recommended and/or suggested dependencies, such
dependencies may be presented to the user through the user
interface 125, and the user may determine whether or not to include
the recommended and/or suggested packages. Alternatively (or
additionally), the development environment 130 may be configured to
automatically add the recommended and/or suggested packages and
enable the user to determine whether or not to remove the
recommended and/or suggested packages that were automatically
added.
[0065] In one implementation, the development environment 130 may
further include an impact analysis engine 135, which may analyze
any changes to the appliance caused by adding and/or removing
software in relation to the appliance, and may further analyze any
changes resulting from the resolution of dependencies for the added
and/or removed software. The impact analysis engine 135 may then
populate the user interface 125 with information describing the
impact of the changes, providing the user with immediate feedback
describing the further development of the appliance. For example,
in one implementation, the impact analysis engine 135 may display a
list of packages added to or deleted from the appliance (including
packages added or deleted in response to the dependency
resolution), an amount of data associated with the software, a
number of pending downloads relating to the software, open source
licenses and associated permissions for the software, or any other
suitable information relevant to the changes to the appliance.
[0066] Additionally, in one implementation, the impact analysis
engine 135 may analyze the changes to the appliance to identify
potential error conditions and provide the user with recommended
options to correct the error conditions. Thus, in response to a
particular change to a appliance causing an error, inconsistency,
or other issue that cannot be fixed automatically, the impact
analysis engine 135 may display a warning within the user interface
125 and may provide a recommendation for the user to manually
correct the issue. For example, if the user adds a package that
declares an open source license in conflict with a license declared
in another package that has already been added to the appliance,
the impact analysis engine 135 may recommend one or more options to
resolve the conflict between the licenses.
[0067] In one implementation, the dependency resolution service 140
may be further configured to provide the user one or more options
to customize various aspects of the appliance. For example, user
interfaces, menus, and other components of the appliance may be
customized around installed and dependent packages (e.g., in
response to the user selecting a user interface or desktop
environment package, "just enough UI" may be installed to support
user interface elements for installed and dependent packages). In
addition, the user may be provided with an undo option to modify
the automatic addition or banning of packages (e.g., the user may
remove dependent packages that were automatically added, override
banned packages to permit their installation, etc.). Further, when
the user removes a package or other component from the appliance,
the dependency resolution service 140 may identify any packages
that were only installed as a dependency of the removed package.
Thus, the dependency resolution service 140 may attempt to
automatically remove unnecessary software in order to maintain a
footprint for the appliance that is as small as possible.
[0068] In one implementation, upon the user having selected the
packages, patterns, or other software components to be included in
the appliance, the user may configure a build for the appliance. In
particular, the user may select a configuration option associated
with the development environment 130 (e.g., a tab, button, or other
option displayed within the user interface 125), which may result
in the development environment 130 loading a configuration
interface within the user interface 125. In one implementation, the
configuration interface may provide one or more options for
establishing settings for a locale (e.g., language, keyboard
layout, region, time zone, etc.), a network (e.g., whether to
configure the network during a first boot, discover settings for
the network automatically using DHCP, manually configure the
network, enable a firewall for the network, etc.), identity
management (e.g., definitions of users, passwords, login
preferences, access controls, etc.), run level (e.g., single user,
multi-user with no network, normal console, graphical login, etc.),
program settings (e.g., applications to start on boot or login,
printer daemons or other system services to be enabled, etc.),
storage and memory (e.g., a default size for a hard disk, memory,
etc.), and/or various other settings. As such, the configuration
interface may provide the user with various options to configure
the build for the appliance, such that a resulting image of the
appliance may be configured to operate in accordance with the
customized settings provided by the user.
[0069] In one implementation, the configuration interface may
further interact with one or more other components described herein
to assist the user in managing certain configuration options. In
particular, in one implementation, the options provided to the user
in the configuration interface may be customized based on metadata
or other information that the repository metadata server 190 and/or
the dependency resolution service 140 discover for the packages or
other software included in the appliance. For example, in response
to a package for a Postgres or a MySQL database being added to the
appliance, the user may be provided with one or more options to
configure specific services associated with the Postgres or MySQL
database package (e.g., a web server appliance may typically
include one or more databases, which may be configured with tables
that are pre-loaded using data identified in a database schema dump
file, with identity management settings or permissions for
different users, etc.).
[0070] Furthermore, in one implementation, the impact analysis
engine 135 may be configured to analyze the configuration settings
established by the user, such that the impact analysis engine 135
may determine whether the configuration settings result in (or
should result in) one or more changes to the appliance. For
example, in response to the user specifying that the network is to
be configured during a first boot, the dependency resolution
service 140 may determine that the "yast2-firstboot" package must
be installed in order for the network to be configured during the
first boot (i.e., the "yast2-firstboot" package may include a first
boot utility that runs after installation completes in order to
guide the user through various steps for configuring a machine). As
such, the impact analysis engine 135 would display a notification
for the "yast2-firstboot" dependency in the user interface 125 and
provide a recommendation for resolving the dependency (e.g.,
displaying an option to add the yast2-firstboot package).
[0071] In one implementation, after the user has configured the
appliance and any software added to the appliance that has
configurable options, the user may personalize the build for the
appliance. In particular, the user may select one or more
personalization options associated with the development environment
130 to load a custom files interface within the user interface 125.
In one implementation, the custom files interface may provide one
or more options for the user to add one or more custom files or
archives to the appliance (e.g., the archives may be in a format
such .tar gz, .tar bz2, .tgz, .tbz, .zip, etc.). As such, the
custom files or archives may be added to the appliance in addition
to any files, packages, or other software that may have been added
through the repository metadata server 190. For example, the user
may upload the custom files or archives from the client device 105,
or the user may identify a location (e.g., a URL) where the custom
files or archives are located. In addition, the user may optionally
specify a target path or directory for the custom files or
archives, as well as identity management settings (e.g., owning
users or groups, permissions, etc.). As such, if the user adds one
or more custom files or archives to the appliance, the custom files
or archives may be copied to the appliance within the specified
target directory after all other software selected for the
appliance has been installed.
[0072] Additionally, in one implementation, one or more of the
personalization options may cause the development environment 130
to load a personalization interface within the user interface 125,
wherein the personalization interface may include one or more
options for the user to personalize the build for the appliance.
For example, the personalization interface may include options for
establishing a graphic to be used as a logo or icon for the
appliance, a background graphic for screens associated with a boot
selection, boot, login, console, desktop, or other screen displayed
when the appliance executes. In one implementation, the user may
select one or more of the logo, background, or other graphics from
an inventory provided in the personalization interface, or the user
may upload graphics to personalize the visual features of the
appliance, wherein the graphics uploaded by the user may optionally
be added to the graphics inventory and made available to other
users of the virtualization environment 120.
[0073] As such, in one implementation, the personalization
interface may be used to extend branding functionality to various
user interfaces in the appliance that include logos or graphics.
For example, the packages, patterns, or other software added to the
appliance may be analyzed to identify where graphics are located or
otherwise referenced, and a template may then be applied to the
identified graphics to enable the graphics to be replaced, deleted,
or otherwise modified. For example, the user may apply custom logos
or graphics to replace one or more of the graphics and essentially
"re-brand" the appliance, or the user may conversely remove one or
more of the identified graphics (e.g., because the user lacks a
license or other authorization to use an embedded trademark, in
which case removal of the graphic trademark may be required).
Furthermore, in cases where the user applies one or more logos or
graphics to create a re-branded appliance, the applied logos or
graphics may be used in distributable media for the re-branded
appliance.
[0074] Furthermore, in one implementation, the virtualization
environment 120 may provide one or more application program
interfaces 150 to enable users, third-party developers, or various
other entities to develop custom plug-ins or applications that can
be used in the virtualization environment 120. The custom plug-ins
or applications may be developed using any suitable programming
language capable of communicating with the virtualization
environment 120 through the application program interface 150. For
example, custom plug-ins or applications may be designed to
configure appliances or specific types of software (e.g., to edit
or configure specific functionality, such as for Apache packages),
to customize appliance creation processes (e.g., to provide social
networking features, such as managing the appliance marketplace),
or to otherwise modify any suitable functionality that may be
provided in the virtualization environment 120. Thus, the one or
more application program interfaces 150 may provide a development
platform for integrating third-party software, as described in
further detail in co-pending U.S. patent application Ser. No.
______, entitled "System and Method for Managing a Virtual
Appliance Lifecycle," filed on an even date herewith, the
disclosure of which is hereby incorporated by reference in its
entirety. As a result, various features and functions associated
with the virtualization environment 120 may be extended through the
one or more application program interfaces 150, thereby leveraging
collective knowledge that may be held within a development
community, whereby the lifecycle of virtual appliances can be
managed within the virtualization environment 120 in a customized
and personalized manner.
[0075] In one implementation, the virtualization environment 120
may further include a build engine 175, which may be configured to
build an image of the appliance that the user created, configured,
personalized, and/or otherwise developed within the development
environment 130. In particular, the development environment 130 may
create an image description for the appliance, wherein the image
description may generally include the packages, patterns, or other
software components to be included in the appliance and the origin
repositories 115 containing the selected software components. In
addition, the image description may optionally further include any
configuration and/or personalization options that the user may have
provided for the software and/or the appliance. The development
environment 130 may then provide the image description for the
appliance to the build engine 175, whereby the build engine 175 may
use the image description to build the image for the appliance.
[0076] In one implementation, the build engine 175 may be invoked
in response to the user selecting an option to build a bootable
image of the appliance, wherein the build engine 175 may then
display one or more options within the user interface 125 to enable
the user to configure the build of the image. For example, the
build options may include an option to specify a name for the
image, which may be displayed on a boot screen for the appliance, a
version number for the Image, and an origin for the appliance
(e.g., a base appliance cloned and derived to create the
appliance). In addition, the build options may include an option to
specify a format for the image, wherein the image may be formatted
as an HDD/USB disk image, an .iso live CD/DVD image, a .vmdk VMware
virtual machine image, and/or an .img Xen virtual machine image,
among other image formats. The user may then select an option to
initiate building the bootable image of the appliance.
[0077] For example, in one implementation, the build engine 175 may
launch one or more virtual machines from a virtual machine pool 170
in response to the user selecting the option to initiate building
the bootable image of the appliance. In particular, the launched
virtual machine may provide a contained virtual environment for
building the image, such that the build engine 175 can
simultaneously and independently build images for more than one
appliance at a particular time (i.e., the build process may operate
at a root user level, whereby building images within contained
virtualized environments may permit multiple images to built
simultaneously using resources that are shared across the
virtualization environment 120). Thus, various different users of
the virtualization environment 120 may initiate one or more builds
at any particular time, wherein the contained build environment may
preserve shared resources of the hosted virtualization environment
120 that may needed for users at various different client devices
105.
[0078] In one implementation, the build engine 175 may be further
configured to provide information relating to a current status for
the build throughout the build process. For example, the build
engine 175 may communicate the current status to the user through
the user interface 125, wherein the current status may include an
elapsed time for the build, a progress meter visually indicating a
degree to which the image has been built or remains to be built,
and a current status of the build process (e.g., "Setting up build
environment," "Preparing system image," "Creating boot ramdisk,"
etc.). In one implementation, the build engine 175 may further
provide links to one or more logs associated with the build
process, including a build log providing information relating to
processes that occur during the entire build, a Kiwi log providing
information relating to creation of a system image, or various
other logs. As such, the one or more logs associated with the build
process may provide the user with a mechanism to debug or otherwise
manage the build process (e.g., if an error should occur during the
process of building the appliance image). Furthermore, in one
implementation, the build engine 175 may provide limited access to
one or more debugging files (e.g., for administrators or other
authorized users), wherein the debugging files may include a
config.xml file, an extra_config.xml file, and an
ensconce_manifest.txt file, among other debugging files.
[0079] In one implementation, the build engine 175 may then store
the completed build of the appliance image in the appliance
repository 180, and the user may be provided with one or more
options to interact with the appliance image. For example, names,
version numbers, image formats, compressed and uncompressed sizes,
build dates, or other information relating to the user's appliance
images may be displayed within the user interface 125. As such, the
user may delete one or more of the images from the appliance
repository 180, publish one or more of the images to the appliance
marketplace, and/or deploy one or more of the images for execution
in one or more runtime environments.
[0080] For example, in one implementation, the user may select one
or more of the appliance images for download to the client device
105, wherein the user may then execute the appliance in an
appropriate runtime environment at the client device 105. In
addition, the user may deploy one or more of the appliance images
to a web server, an Amazon Elastic Compute Cloud (EC2), or any
other environment suitable for hosting or otherwise executing the
appliance. Alternatively (or additionally), the user may execute
one or more of the appliances in a runtime environment 160 hosted
at the virtualization environment 120. For example, in response to
the user selecting an option to execute the appliance in the hosted
runtime environment 160, a virtual machine may be launched from the
virtual machine pool 170 to create a contained runtime environment
160 for executing the appliance. As such, the user may execute the
appliance within the contained runtime environment 160 in a manner
that preserves resources that are shared across the virtualization
environment 120. In addition, the hosted runtime environment 160
may include a monitoring engine 165 having one or more appliance
management utilities (e.g., the appliance may be executed in the
hosted runtime environment 160 to determine whether the appliance
functions properly or was correctly built, to identify un-utilized
or under-utilized files that can be removed from the appliance, to
characterize certain processes that occur during execution of the
appliance, etc.).
[0081] Additionally, in one implementation, the virtualization
environment 120 may further include a support engine 155 configured
to manage updates and supportability for appliances that are
managed in the virtualization environment 120. For example, as
noted above, a user may install various packages, patterns, or
other software components to be included in an appliance, wherein
the selected software components may be installed from one or more
origin repositories 115. Thus, in response to the repository
metadata server 190 detecting an upstream update to one or more of
the origin repositories 115 (e.g., if local digests and current
digests for the origin repositories 115 do not match), an update
server coupled to the support engine 155 may identify any
particular packages, patterns, or other software components that
have been updated in the updated origin repositories 115. The
update server may then further identify any of the managed
appliances that include the particular updated packages, patterns,
or other software components and generate a workflow for updating
the identified appliances.
[0082] For example, in one implementation, the update workflow may
generally include sending a notification to the users that created
the appliances, indicating that updated versions have been released
for one or more of the software components installed for the
appliance. In particular, the workflow may include rebuilding
images for the appliances to install the updated software
components, and may further include managing an integration test to
ensure that the rebuilt images function or otherwise behave
correctly with the updated software components installed. For
example, the updated software components may have different
features or behaviors from the original versions, which may not
necessarily integrate with other components installed for the
appliance, whereby the integration test may be performed to ensure
that the appliance continues to function correctly in response to
the updated software components being installed. In response to the
integration test being performed and the user determining that the
appliance no longer functions correctly, the workflow may further
enable the user to implement any changes to the appliance that may
be needed to accommodate the updated software components, and
further to enable the user to perform another integration test. As
such, one or more integration tests may be performed until the user
determines that the appliance functions correctly with the updated
software components installed.
[0083] In one implementation, in response to the user determining
that the rebuilt appliance functions correctly with the updated
software components installed, the rebuilt appliance incorporating
the updated software components may be released. As such,
subsequent users that clone or otherwise deploy the rebuilt
appliance may receive current versions for all of the installed
software components. In one implementation, the update server may
further generate one or more updates corresponding to the rebuilt
appliance, wherein the updates may be released to earlier versions
of the rebuilt appliance. For example, the earlier versions of the
rebuilt appliance may be configured to periodically contact the
virtualization environment 120 to determine whether any updates are
available, and the update server may release the updates to the
earlier versions of the appliances to install the updated software
components. In addition, in one implementation, the rebuilt
appliance and/or the updates generated on the update server may be
released with one or more access control mechanisms, which may
define particular users, groups of users, or other criteria that
control the release.
[0084] In one implementation, the support engine 155 may further
include a support analysis manager configured to analyze
supportability and service support requests for one or more of the
appliances managed in the virtualization environment 120. For
example, as noted above, the virtualization environment 120 may
provide various JeOS templates that may be cloned to create a base
appliance providing a starting point for developing a virtual
appliance, wherein one or more of the JeOS templates may be based
on an operating system distributed by the provider of the
virtualization environment 120. In addition, the virtualization
environment 120 may further enable an independent software vendor
(ISV), developer, or any other user to install various packages,
patterns, or other software components to create an appliance that
customizes the base operating system distribution for a particular
application. Consequently, although the ISV, developer, or other
user may have knowledge relating to the particular application for
which the appliance has been customized, in many instances the ISV,
developer, or other user may lack detailed knowledge relating to
the underlying operating system distribution. As such, when
customers or other users of the appliance encounter problems with
the appliance (e.g., because the appliance crashes or functions
less than optimally), the ISV, developer, or other user may lack
sufficient knowledge to provide support for the problems,
especially when the problems involve the operating system
distribution or software components other than the particular
application for which the appliance has been customized.
[0085] Accordingly, in one implementation, the support analysis
manager may be configured to analyze various appliances and
generate support statements indicating whether the provider of the
virtualization environment 120 can provide support for certain
components that have been installed and/or certain problems that
may be encountered for the appliances. For example, as noted above,
one or more of the JeOS components that can be installed for an
appliance may be include a Linux operating system distribution,
which typically has a modular architecture based on open source
code that can be stripped and modified as needed. Thus, if the
support analysis manager analyzes a particular appliance and
determines that the appliance includes a modified kernel, new
drivers, or other customizations that substantially alter the base
operating system distribution, the support statement may indicate
that the provider of the virtualization environment 120 cannot
support problems relating to the modified kernel, new drivers, or
other customizations. However, if the appliance merely includes a
different version of a particular package than a corresponding
version of the package that is shipped with the base operating
system distribution (i.e., the core components of the base
operating system remain intact), the support statement may indicate
that problems relating to the base operating system distribution
can be supported, while any components of the appliance that have
been substantially altered cannot be supported. Thus, the support
analysis manager may determine whether any components installed for
the appliance that are shipped by the provider of the
virtualization environment 120 remain intact, wherein the
determination may be analyzed in view of one or more heuristic
rules to generate the support statement.
[0086] Furthermore, in one implementation, one or more of the
packages, patterns, or other software components that can be added
to a custom appliance may originate from trusted partners or other
entities known to the provider of the virtualization environment
120. For example, as noted above, a trust metric may be provided
for one or more of the origin repositories 115 and/or the packages,
patterns, or other software components contained therein, wherein
the trust metric may indicate that the relevant origin repositories
115 and/or packages, patterns, or other software components
originate from the trusted partners or other known entities. Thus,
in response to the support analysis manager determining that the
provider of the virtualization environment 120 cannot support
certain components installed for an appliance, the support analysis
manager may further determine whether the trusted partners or other
known entities can support such components. As such, in one
implementation, the support statement generated by the support
analysis manager may identify any relevant trusted partners or
other known entities that can provide the appropriate support.
Furthermore, the support analysis manager may analyze the appliance
in a similar manner as described above to determine whether the
components supported by the trusted partners or other known
entities have been substantially altered to the extent that support
cannot be provided. For example, in one implementation, a
cryptographic key (e.g., a PGP certificate) may be used to indicate
whether a particular package, pattern, or other software component
is "trusted," wherein the cryptographic key for the trusted package
may be configured to break or otherwise cease to be intact in
response to substantial alterations. Thus, the support analysis
manager may be configured to determine whether the cryptographic
key or other relevant data remains intact, wherein the
determination may then be considered in combination with one or
more heuristic rules to generate the support statement.
[0087] According to one aspect of the invention, FIG. 2 illustrates
a flow diagram of an exemplary method for creating virtual
appliances in a system for managing a virtual appliance lifecycle.
In particular, the method illustrated in FIG. 2 may begin in an
operation 210, wherein a user at a remote client device establishes
a connection to a system for managing a virtual appliance lifecycle
(e.g., the system illustrated in FIG. 1 and described in further
detail above). The user may then include launch a development
environment associated with the virtual appliance management system
in order to create one or more virtual appliances. In one
implementation, upon launching the development environment,
operation 210 may further include receiving information from the
user that identifies a base appliance to be created in the
development environment.
[0088] For example, in one implementation, the information received
from the user in operation 210 may include a selection of a
predefined JeOS template (e.g., a template for a virtual appliance
having a minimal operating system footprint). Alternatively, the
information received from the user may include a request to clone
an appliance that the user previously developed in the development
environment, a request to clone an appliance from an appliance
marketplace (e.g., appliances that other users have previously
developed and published to the appliance marketplace), or an
imported configuration file that includes an image description for
an appliance (e.g., directory structures, configuration scripts,
packages, software, or other information describing an appliance
image). In one implementation, in response to receiving the
information identifying the base appliance in operation 210, the
base appliance in may be created and loaded within an editing
interface in an operation 220. As such, the user may then interact
with various features provided in the editing interface to further
develop the base appliance within the development environment.
[0089] In one implementation, once the base appliance has been
loaded for editing within the editing interface in operation 220,
the user may access a repository metadata server to add, remove,
update, or otherwise query information associated with one or more
origin repositories that include packages, patterns, or other
software that can be added to and/or banned from the base
appliance. As such, in one implementation, an operation 230 may
include determining whether a request has been received from the
user to add, remove, update, or otherwise query information
associated with one or more of the origin repositories. In response
to operation 230 determining that the user has provided a request
relating to one or more of the origin repositories, an update for
the origin repositories identified in the request may be initiated
in an operation 235.
[0090] For example, if the request received from the user
identifies a new origin repository to be added to the repository
metadata server, operation 235 may include prompting the user to
define a name and/or a location for the origin repository (e.g., a
URL or other location for contacting the new origin repository).
The repository metadata server may then use the location
information provided by the user to contact the new origin
repository and retrieve metadata describing the new origin
repository and any packages, patterns, or other software that may
be contained therein. Thus, in one implementation, updating the
origin repositories in operation 235 may include inserting the
metadata retrieved from the new origin repository into a local
repository database coupled to the development environment. As
such, the user may then query the local repository database to
request information relating to the new origin repository, any
packages, patterns, or software contained therein, or any other
information that can be derived from the metadata retrieved from
the new origin repository (e.g., dependency relationships for the
packages, patterns, or other software in the origin repository). In
one implementation, the packages, patterns, or other software in
the new origin repository may be downloaded to one or more locally
stored cached repositories, thereby enabling the user to interact
with the packages, patterns, or other software contained in the new
origin repository through the development environment.
[0091] Furthermore, in one implementation, the request received
from the user in operation 230 may include a selection of one or
more of the origin repositories available through the repository
metadata. For example, the user may query the local repository
database to request one or more of the origin repositories
available through the repository metadata server (including any
repositories that may have been added by the user), or the user may
query the repository database to search for one or more of the
origin repositories having a particular name, a particular type, or
particular software (e.g., the user may enter a search string of
"Apache" to search the repository database for origin repositories
that contain one or more Apache packages). As such, in response to
the user selecting one or more origin repositories in operation
230, the selected origin repositories may be updated in operation
235. For example, the repository metadata server may contact the
selected origin repositories to determine whether any the origin
repositories are outdated (e.g., an origin repository may be
outdated if a digest describing a current state of the origin
repository does not match a digest that was previously retrieved
from the origin repository). Thus, if one or more of the selected
origin repositories are outdated, operation 235 may include
updating the repository database with current metadata for the
outdated origin repositories, downloading updated packages,
patterns, or other software from the outdated origin repositories
to the cached repositories, and/or building updated resolution data
files for the packages, patterns, or other software contained in
the outdated origin repositories.
[0092] In one implementation, the origin repositories requested by
the user in operation 230 may then be used to provide a source of
software for the virtual appliance. As such, an operation 240 may
include receiving one or more requests from the user, wherein the
requests may identify one or more packages, patterns, files, or
other software components to be added to and/or banned from the
base appliance previously loaded in operation 220. For example, the
user may browse the repository database according to one or more
software groups to locate one or more packages, patterns, or other
software components having specific functionality (e.g., the
software groups may be organized based on functionality relating to
operating systems, development, games, graphics, multimedia, office
suites, systems, servers, networking, desktop environments,
etc.).
[0093] In one implementation, the requests may further include
criteria for searching the repository database for certain
packages, patterns, or other software components that match the
search criteria. As such, the request received from the user in
operation 240 may identify one or more packages, patterns, or other
software to be added to and/or banned from the base virtual
appliance. In one implementation, the description provided above in
relation to FIG. 1 may provide further detail regarding the manner
in which the user can browse and/or search the repository database
to manage software for the appliance (e.g., results of the browsing
and/or searching may be filter according to the origin repositories
requested in operation 230, all origin repositories available
through the repository metadata server, popularity, trust, or other
metrics, etc.).
[0094] In one implementation, in response to the user selecting a
particular software component to be added to and/or banned from the
appliance in operation 240 (e.g., a package, pattern, etc.), one or
more dependencies for the selected component may be resolved in an
operation 250. In particular, as described in further detail above,
any particular package, pattern, or other software component may be
associated with metadata describing dependencies for the particular
package, pattern, or other software component. For example, in one
implementation, the dependencies may include one or more related
components that are required for the selected component to operate
properly, related components that are recommended (but not
required) for the selected component to operate optimally, related
components that are suggested (but not required) to enhance the
functionality of the selected component, and/or related components
that conflict (and thus cannot be installed) with the selected
component. As such, in response to the user selecting a particular
package, pattern, or other software for the appliance, the metadata
associated with the selected component may be scanned in operation
250 to determine whether or not the selected component has one or
more dependencies.
[0095] In one implementation, in response to determining that the
selected component has one or more dependencies in operation 250, a
resolution request may then be initiated in an operation 255. For
example, a dependency resolution service (e.g., as described above
in relation to FIG. 1) may use the metadata describing the origin
repository associated with the selected component, and may further
use resolution data constructed by the repository metadata server,
in order to resolve the dependencies for the selected component.
For example, in one implementation, operation 255 may include the
dependency resolution service invoking an active resolution daemon
to identify any software components that may be dependent,
recommended, suggested, and/or conflicting with respect to the
selected component. In particular, the active resolution daemon may
scan one or more resolution graphs to identify annotations and/or
metadata that explicitly describe dependencies for the component,
in addition to any pre-installation scripts, post-installation
scripts, and/or content associated with the component to
introspectively identify further dependencies (e.g., based on
documentation included in the package). Thus, in response to
resolving the dependencies for the particular component, operation
255 may include automatically adding any required dependencies to
the appliance, and may further include automatically removing any
conflicting dependencies from the appliance. Additionally, in one
implementation, any recommended and/or suggested dependencies may
be automatically added in operation 255, or the user may optionally
be notified of such dependencies to provide the user with control
over whether or not to include the recommended and/or suggested
dependencies.
[0096] In one implementation, once the user has selected one or
more software components to be added to and/or banned from the
appliance and any dependencies for the particular components have
been resolved, an impact analysis may be performed in an operation
260 to identify any resulting changes to the appliance. For
example, operation 260 may include generating a visual notification
informing the user of the impact (e.g., displaying a list of
packages added to or deleted from the appliance, a data impact on
the appliance in terms of megabytes, gigabytes, etc.).
Additionally, the impact notification may provide the user with one
or more undo capabilities and/or error correction capabilities. For
example, the undo capabilities may enable the user to remove one or
more packages that were added to the appliance, override a ban
placed on one or more packages due to a conflicting dependency,
override the removal of a package from the appliance due to a
conflicting dependency, or otherwise undo one or more changes to
the appliance. In addition, if a particular change to an appliance
causes an error, inconsistency, or other issue that cannot be
resolved automatically, the error correction capabilities may
include displaying a warning or other notification together with
one or more options or recommendations to correct the issue.
[0097] In one implementation, after the user has selected the
packages, patterns, or other software components to be included in
and/or banned from the appliance, the user may configure and
personalize an image to be built for the appliance in an operation
270. For example, in operation 270, the user may configure the
build to establish settings relating to locale, network
configuration, identity management, login preferences, database
configuration, storage and memory, or various other settings.
Additionally, the user may personalize the build for the image in
operation 270 to select logos, icons, backgrounds, or features to
personalize the image. In one implementation, operation 270 may
then include building a bootable image of the appliance, wherein
one or more virtual machines may be launched to create a contained
build environment within which the appliance image may be
generated. The user may then deploy the appliance image for
execution in one or more runtime environments (e.g., the user may
download the appliance image for execution in a local runtime
environment, load the appliance image within a hosted runtime
environment, deploy the appliance image to a cloud computing
environment, etc.). As such, the method illustrated in FIG. 2 and
described herein may provide a simple and repeatable process for
managing the creation of virtual appliances.
[0098] According to one aspect of the invention, FIG. 3 illustrates
a flow diagram of an exemplary method for building virtual
appliances in a system for managing a virtual appliance lifecycle.
In particular, one or more users may create, configure,
personalize, and otherwise develop one or more appliances in a
virtualization environment, as described in further detail above.
In one implementation, the virtualization environment may further
include a build engine configured to build images (e.g., operating
system images) of appliances that the users develop in the
virtualization environment. For example, a typical process for
installing an operating system generally includes installing
various individual software components from a particular
installation source, often requiring further manual user
intervention. In contrast, the appliance images built using the
build engine may generally include one file that encapsulates a
complete operating system installation and/or various
configurations for particular tasks or software components. As
such, the appliance images may provide fully configured systems
that can be executed in response to being deployed to an
appropriate storage medium.
[0099] In one implementation, the method illustrated in FIG. 3 may
be initiated in response to receiving a build request from a user
in an operation 310. In particular, the build request may generally
include an image description that provides various parameters
defining the virtual appliance, including any software that the
user has added to the appliance and any configurations and
personalizations that the user has defined for the appliance and/or
the software. Additionally, in one implementation, the build
request may further include a format that the user has selected for
the image (e.g., an HDD/USB disk image, an .iso live CD/DVD image,
a .vmdk VMware virtual machine image, an .img Xen virtual machine
image, etc.). The virtualization environment may then validate the
image description to determine whether the image description
includes the appropriate information needed to build the image
(e.g., valid origin repositories and/or other sources for the
software, valid directory structures, and configurations for the
software, etc.). In one implementation, if the image description
contains any errors or other invalid information, the
virtualization environment may attempt to identify the errors and
provide the user with warnings and/or options for resolving the
errors (e.g., providing an option to install a specific package
needed for a particular software configuration).
[0100] In one implementation, in response to the image description
being validated, the virtualization environment may then launch the
build engine in an operation 320. In particular, to launch the
build engine in operation 320, the virtualization environment may
load a virtual machine from a virtual machine pool to provide a
contained environment within which the image may be built. For
example, in one implementation, a snapshot of the build engine may
be taken when the build engine is ready to build an image (e.g.,
following any initialization processes such as locating required
build services, setting up logging mechanisms, etc.). As such, a
subset of the virtual machines in the virtual machine pool may be
in a pre-configured state corresponding to the snapshot of the
initialized build engine, such that the pre-configured virtual
machines may be used to reduce an amount of time needed to launch
the build engine and initiate building the image. Moreover,
providing the virtual machine within a contained environment may
enable the build engine to build various different images
corresponding to various different appliances in a substantially
simultaneous and independent manner. For example, any particular
build may generally begin with creating an encapsulated physical
extend reachable at a root level, whereby contained build
environments may provide independent physical extends for various
simultaneous builds at the root level. Thus, the contained build
environment launched in operation 320 may preserve resources that
can be shared by various users across the virtualization
environment.
[0101] In one implementation, the build engine may then build the
appliance image within the contained build environment in an
operation 330a. In particular, in operation 330a, the build engine
may analyze the image description previously received in operation
310 and create a physical extend based on the image description
within the contained build environment. For example, the physical
extend may generally include a file system layout having a root
directory and one or more sub-directories within the root
directory. In one in implementation, the file system layout may
further include a file layout identifying locations within the file
system where one or more files are to be installed. After the
physical extend and the corresponding file system layout has been
created, operation 330a may further include creating the image (or
logical extend) from the physical extend. For example, to create
the image of the appliance, the build engine may retrieve the files
to be installed within the directories and sub-directories of the
file system (e.g., from the origin repositories, cached
repositories, or other sources). The build engine may then copy the
files to the image in accordance with the file system layout
defined in the physical extend (e.g., as described in further
detail in co-pending U.S. patent application Ser. No. ______,
entitled "System and Method for Efficiently Building Virtual
Appliances in a Hosted Environment," filed on an even date
herewith, the disclosure of which is hereby incorporated by
reference in its entirety). As such, the image may include one file
encapsulating the files corresponding to the software selected for
the appliance, with the files being organized within the file
system layout, and with the image being further configured and
personalized according to any parameters that the user may have
provided for the appliance.
[0102] In one implementation, operation 330a may further include
the build engine formatting the image in accordance with the format
that the user selected for the image. For example, in one
implementation, the user may select a disk image format or an .iso
image format, in which case the build engine may format the image
for deployment onto physical media that can be executed in a
suitable runtime environment. The disk image or iso image may then
be copied to a hard disk, USB stick, CD, DVD, or other physical
media that can be mounted or otherwise installed for execution on a
system. Alternatively, the disk image or .iso image may be copied
to an OEM virtual disk that can be pre-loaded or otherwise
installed for execution on the system. In another example, the user
may select a .vmdk virtual disk format, in which case the build
engine may format the image for deployment within a VMware virtual
machine that can execute the image, or the user may select an .img
virtual image format, in which case the build engine may format the
image for deployment within a Xen hypervisor environment that can
execute the image. As such, the build engine may support various
formats for the image, which can then be deployed and executed
within various different runtime environments.
[0103] In one implementation, the build engine may further be
configured to communicate with the support engine described in
further detail above in connection with FIG. 1, wherein an
operation 330b may include the build engine invoking the support
engine to analyze supportability for the appliance corresponding to
the current build. In particular, the support engine may include a
support analysis manager configured to analyze the supportability
for the appliance, whereby the support analysis manager may
generate a support statement in operation 330b to indicate whether
support can be provided for certain components that have been
installed for the appliance. For example, in one implementation,
the virtualization environment may provide one or more guidelines
regarding supportability for various packages, patterns, and other
software components that a provider of the virtualization
environment and trusted partners and other entities known to the
provider can support (e.g., the supportability guidelines may
describe certain changes or modifications that can or cannot be
implemented for the supported components). As such, operation 330b
may be integrated within the build process illustrated in FIG. 3
and described herein, wherein operation 330b may include analyzing
the appliance to ensure that the appliance adheres to the
supportability guidelines.
[0104] For example, in one implementation, the support analysis
manager may be configured to analyze each of the packages,
patterns, and other software components installed for the appliance
to generate the support statement indicating whether each of the
installed packages, patterns, and other software components can be
supported by the virtualization environment provider, the trusted
partners, or the other known entities. In particular, rather than
providing a binary "yes" or "no" describing whether or not support
can be provided, the support analysis manager may analyze all of
the installed packages, patterns, and other software components in
view of various heuristic rules to validate whether the appliance
adheres to the supportability guidelines (e.g., the heuristic rules
may specify permitted variants for packages, patterns, and other
software components available for installation in a particular
appliance). As such, in one implementation, the support analysis
manager may analyze the appliance subject to the current build
against the permitted variants to generate the support statement,
which may indicate whether particular ones of the installed
packages, patterns, and other software components can or cannot be
supported.
[0105] In one implementation, to analyze the appliance against the
permitted variants, the support analysis manager may be configured
to retrieve metadata for the installed packages, patterns, and
other software components, wherein the heuristic rules may express
the supportability guidelines in terms of the metadata. Thus, any
particular package available for installation in an appliance may
be associated with metadata describing the package, such that one
or more heuristic rules may express supportability guidelines for
the package in terms of the metadata describing the package. For
example, the package may include several files and core components
that are critical to the package functioning correctly, and may
further include several files that provide documentation for the
package. As such, the heuristic rules for the package may indicate
that the documentation files can be removed from the package with
impacting the supportability of the package, while further
indicating that removing or modifying the critical files and/or
core components render the package unsupportable. Accordingly, in
one implementation, the support analysis manager may analyze the
metadata for any supportable software components installed for the
appliance in view of any associated heuristic rules to generate the
support statement indicating whether or not the supportable
software components have been modified to the extent of being
rendered unsupportable.
[0106] In one implementation, the build engine may be further
configured to monitor the build for the image during operations 320
and 330a-b, such that an operation 340 may include generating one
or more logs describing the build process. Thus, in one
implementation, the build engine may display the current status for
the build during operations 320 and 330a-b within a user interface.
For example, the current status may include information such as an
elapsed time for the build, a progress bar indicating a degree of
completion for the build, and a current process being executed
(e.g., "Setting up build environment," "Preparing system image,"
"Copying packages," "Creating boot ramdisk," etc.). As such, the
build engine may continually record the current status for the
build within one or more build logs and then finalize the build
logs in operation 340 in response to build operations 320 and
330a-b completing. The build logs may therefore provide a record
describing the various processes executed during the build, in
addition to the support statement, and any errors, problems, or
other status changes that occurred during the build. For example,
in one implementation, the build logs generated in operation 340
may include a Kiwi log describing creation of the image, a
debugging log referencing files or other information used to create
the image, or any other logs that provide a record of information
relating to a build. Thus, the user may reference the build logs to
debug or otherwise review the build process.
[0107] In one implementation, in response to the build engine
successfully creating the image of the appliance, an operation 350
may include determining whether the appliance was created from a
cloned appliance. For example, operation 350 may determine that the
appliance was created from a cloned appliance if a user created the
appliance from a JeOS template provided in the virtualization
environment, an appliance previously created by the user in the
virtualization environment, or an appliance published in the
appliance marketplace. As such, in response to determining that the
appliance was created from a cloned appliance, the image created
for the appliance may be analyzed in an operation 360 to construct
or otherwise update a family tree for the cloned appliance, as
described in further detail below. On the other hand, in response
to determining that the appliance was not created from a cloned
appliance (e.g., if the user created the appliance from an imported
Kiwi configuration file containing an appliance description),
processing may proceed directly to an operation 370 to handle
publishing the image created for the appliance, also described in
further detail below.
[0108] In one implementation, as noted above, a family tree for the
cloned appliance may be constructed or otherwise updated in
operation 360 in response to determining that the appliance was
created from a cloned appliance. In particular, as discussed above
in connection with FIG. 1, the virtualization environment may
provide an appliance marketplace within a panel or other interface
that can be browsed and/or searched for various appliances that
users of the virtualization environment have made publicly
available. Thus, users of the virtualization environment may
interact with the appliance marketplace to view information
relating to the public appliances (e.g., list of packages, file
system layouts, installed files, enabled services, amounts of
gigabytes and/or megabytes for the appliances, etc.). In addition,
the users may clone any of the appliances in the appliance
marketplace and implement changes to the cloned appliance, such
that various appliances may be derived from one other. As such,
various relationships between the cloned appliance and the image
built in operation 330a may be analyzed to construct the family
tree in operation 360.
[0109] For example, in one implementation, the family tree may
represent links between the cloned appliance and any appliances
derived from the cloned appliance. In particular, the family tree
may include a hierarchy having a parent node to represent the
cloned appliance, with the parent node having one or more children
nodes, grandchildren nodes, or other descendant nodes representing
the derived appliances. As such, a relationship between the parent
node and the children nodes may be represented with a branch
connecting the parent node to the children nodes. In addition, at
varying levels of the hierarchy, different appliances derived from
the same cloned appliance may be represented as sibling nodes.
Thus, the family tree may arrange various nodes in a hierarchy to
represent the various relationships for cloned and derived
appliances, while changes or other differences among the related
appliances may be analyzed to enhance the information provided in
the appliance marketplace. Thus, operation 360 may include
analyzing the related appliances in the family tree according to
one or more data mining algorithms, heuristic algorithms, and/or
other analytic techniques to derive knowledge relating to the
changes or other differences among the related appliances.
[0110] As such, in one implementation, operation 360 may include
analyzing the appliances linked to one another in the family tree
may to organize the appliances into one or more clusters (e.g.,
related appliances incorporating similar changes to a cloned
appliance). As such, the clusters may then be further analyzed to
identify one or more "best changes," "recommended changes," and/or
"popular changes." In addition, information relating to the changes
between the various appliances in the family tree may be
represented along branches between the nodes of the hierarchy
(e.g., indicating that a derived appliance has changed background
graphics, disabled certain services, added and/or removed certain
packages, or made other changes to the cloned appliance). Thus, in
one implementation, the branches of the family tree may be
associated with metadata describing the changes or other
differences from the parent nodes to their respective children
nodes, while each node in the hierarchy nodes may be associated
with further metadata describing the respective node's cluster
membership, popularity metric, trust metric, or other descriptive
information. Furthermore, in one implementation, sibling nodes
(i.e., appliances derived from the same cloned appliance) may be
analyzed to derive further knowledge relating to the family tree
(e.g., if various different users clone a particular appliance and
make similar changes, a developer of the cloned appliance may be
interested in determining whether or not those changes should be
applied to the cloned appliance).
[0111] Thus, in one implementation, constructing the appliance
family tree in operation 360 may provide various forms of knowledge
that may be used to manage the lifecycle of the appliances in the
family tree, including changes that should be made to subsequent
versions of an appliance, trends or patterns for creating new base
products, and trends or patterns for modifying behavior or
functionality in the virtualization environment, among other
things. For example, any users browsing and/or searching the
appliance marketplace may view the changes between parent nodes,
children nodes, sibling nodes, and other related nodes to
distinguish the different yet related appliances (e.g., a user may
consider a particular sibling node's popularity metric to
distinguish various sibling nodes that incorporate the same "best
changes"). In another example, changes implemented in derived
appliances may be used to update or otherwise optimize earlier
generations of the appliances (e.g., the "best changes," "popular
changes," "recommended changes," or other changes may be assembled
into a package that can update earlier generations and apply the
relevant changes). In still another example, the changes may be
further analyzed to provide feedback that can be used to optimize
or otherwise modify behavior or functionality of the virtualization
environment (e.g., if many users are employing certain
configuration options when configuring appliances, while a smaller
number users employ other configuration options, such knowledge may
be used to make informed choices regarding the configuration
options to be presented within a default configuration interface
versus an advanced configuration interface).
[0112] In one implementation, if the appliance was not derived from
a cloned appliance, or alternatively in response to successfully
constructing the family tree in operation 360, publication of the
image may be managed in an operation 370. In particular, a
publication prompt may be presented to the user that created the
appliance, wherein the publication prompt may enable the user to
establish whether or not the image of the appliance is to be made
publicly visible. Thus, in response to the user indicating that the
appliance is to be published, the image may be added to the
appliance marketplace in an operation 380, thereby enabling other
users to view and optionally clone the published appliance, and the
image may further be added to the user's appliances in an operation
390. Alternatively, in response to the user indicating that the
appliance is not to be published, the image may only be added to
the user's appliances in operation 390 (i.e., the image will be
visible to the user that created the appliance, but will not be
visible to other users of the virtualization environment). However,
in one implementation, the image of the unpublished appliance may
be made visible to limited users (e.g., other users in a friend
list for the user, other users that created earlier generations of
the unpublished appliance, other users that created appliances in
the same family tree as the unpublished appliance, etc.).
[0113] As such, in one implementation, completed builds of
appliance images may be persistently stored in one or more of the
user's appliances and the appliance marketplace, and the user may
be provided with various options for managing the appliance images.
For example, names, version numbers, image formats, compressed and
uncompressed sizes, build dates, architectures, installed packages,
and/or other information describing the appliance images may be
displayed within a user interface. Furthermore, the user may
optionally delete any, download, clone, publish, deploy, or
otherwise manage any of the images.
[0114] According to one aspect of the invention, FIG. 4 illustrates
a flow diagram of an exemplary method for testing virtual
appliances in a runtime environment. In particular, the
virtualization environment may generally store completed images
built for various appliances, wherein users may execute one or more
of the images in a runtime environment hosted at the virtualization
environment. Alternatively (or additionally), users may upload one
or more images to the virtualization environment and execute the
uploaded images in the runtime environment (i.e., images may be
executed using the runtime environment hosted at the virtualization
environment, regardless of whether or not the image was built
within the virtualization environment). In one implementation, the
method illustrated in FIG. 4 may be initiated in response to
receiving a test request from a user in an operation 410. In
particular, the test request may generally include an identifier
for an image to be executed in the hosted runtime environment, and
may optionally further include one or more parameters for executing
the image (e.g., user name and password, enabled and disabled
services, etc.).
[0115] In one implementation, an operation 420 may include the
virtualization environment launching the runtime environment in
response to receiving the test request in operation 410. In
particular, the virtualization environment may load a virtual
machine from a virtual machine pool to provide a contained runtime
environment for executing the image. For example, one or more
snapshots of the runtime environment may be taken when the runtime
environment is ready to execute one or more images having typical
runtime characteristics (e.g., default services enabled and/or
disabled, network configurations, etc.). As such, a subset of the
virtual machines in the virtual machine pool may be in a
pre-configured state corresponding to the snapshots of the
initialized runtime environment, such that the pre-configured
virtual machines may be used to reduce an amount of time needed to
launch the runtime environment and initiate executing the image.
Moreover, providing the virtual machine within a contained
environment may permit various different images corresponding to
various different appliances to be executed in a substantially
simultaneous and independent manner. For example, any particular
image may generally be executed at a root level, whereby contained
runtime environments may permit different images to be executed
simultaneously at the root level. Thus, the contained runtime
environment launched in operation 420 may preserve resources that
can be shared by various users across the virtualization
environment.
[0116] In one implementation, the image may then be executed within
the contained runtime environment in operation 420. In particular,
one of the pre-configured virtual machines having similar runtime
characteristics as the image identified in the execution request
may be loaded, such that the image may be executed within the
appropriate pre-configured virtual machine in operation 420.
Alternatively, in one implementation, multiple virtual machines may
be loaded for a particular image to provide independent runtime
environments for the image (e.g., enabling the user to implement
different configurations, operational characteristics, or other
aspects of the image in a parallel and/or independent manner). As
such, in one implementation, the image may be booted within a guest
operating system associated with the contained runtime environment,
wherein the user can reconfigure the appliance, install and/or
uninstall additional software, or execute any other task that the
image supports at runtime without having to download or otherwise
deploy the image to another runtime environment.
[0117] In one implementation, the user may create a "snapshot" of
the image at any point during the execution of the appliance in
operation 420. As such, an operation 430 may include continuously
or intermittently determining whether a request to create a
snapshot of the appliance was received. In response to determining
that a snapshot request was received, the snapshot may then be
captured in an operation 440. The snapshot may generally capture an
entire state of the image at any particular point, wherein the
snapshot may then be used in various ways to manage the appliance.
For example, the user may capture the snapshot after performing
various tasks while executing the image in the runtime environment,
such that the snapshot can subsequently be loaded to rebuild the
image in a state corresponding to the captured snapshot (e.g., the
user may capture a snapshot prior to performing a particular task
likely to cause errors or other problems, such that the snapshot
can be used to restore the image to the pristine state from prior
to when the particular task was performed). In another example, a
state of the image may be captured in response to the image
successfully booting to provide a snapshot that can be executed
without requiring a boot. As such, a state of the image may be
captured at any particular time, such that the appliance can be
restored to a runtime state corresponding to the snapshot.
[0118] Furthermore, in one implementation, the snapshot capturing
capabilities provided in operation 440 may be used to manage
various aspects of the appliance lifecycle. For example, the
snapshot may correspond to a crashed state for the appliance,
whereby the snapshot can then be provided to an appropriate entity
for debugging the appliance in the crashed state. In another
example, if the image is to be deployed onto a particular storage
medium, the runtime environment may be configured to resize the
image to ensure that the image can fit on the storage medium (e.g.,
if a hard disk has a limited amount of available space, the image
may be resized to fit within the available space). Thus, in one
implementation, various snapshots corresponding to different
optimizations performed for different runtime environments may be
taken for the image, wherein the various snapshots may be built
into the image, thereby enabling the image to be deployed to a
particular runtime environment using an appropriate one of the
various snapshots (e.g., hardware for a particular runtime
environment may be probed to determine a configuration for the
image, such that a snapshot of the configured image may provide a
suitable state for executing the image in that runtime
environment).
[0119] In one implementation, an operation 450 may include
continuously or intermittently determining whether the user has
provided an indication that the user has completed testing the
image. Thus, testing the image may include iteratively repeating
operations 420 through 450 until the user provides an indication
that the user has completed testing the image. In one
implementation, the iterative execution of the image may enable the
user to capture various snapshots for the image at different points
in time, wherein the various snapshots may further enable
management of the appliance lifecycle. For example, a first
snapshot may be captured at a first point in time, and a second
snapshot may be captured at a subsequent point in time, wherein any
changes or other differences between the first snapshot and the
second snapshot may be attributed to any runtime activity that
occurred between the first point in time and the subsequent point
in time (e.g., an installer program may be executed after capturing
the first snapshot, and the second snapshot may then be captured to
identify particular changes that are caused by the installer
program). In this manner, different snapshots taken at different
points in time may be compared to one another to provide feedback
for auditing one or more specific activities (e.g., simplifying
development for installer scripts or cross-platform software,
generating encryption activity records for security management,
tracking relationships between different versions of an appliance,
etc.).
[0120] In one implementation, in response to the user indicating
that the execution context for the image has completed, one or more
runtime logs may be displayed in an operation 460. For example, as
noted above, the hosted runtime environment may include a
monitoring engine having one or more appliance management utilities
for monitoring the execution that occurs during operation 420
(e.g., as described in further detail in co-pending U.S. patent
application Ser. No. ______, entitled "System and Method for
Inspecting a Virtual Appliance Runtime Environment," filed on an
even date herewith, the disclosure of which is hereby incorporated
by reference in its entirety). The appliance management utilities
may therefore be used to generate the runtime logs displayed in
operation 460, wherein the runtime logs may identify any files that
were modified, unused, or underused during execution of the
appliance image (e.g., whether and/or to what certain packages,
patterns, components, or other aspects of the appliance were used
or changed during the execution).
[0121] In particular, in one implementation, the runtime logs may
include any files that were modified subsequent to the image being
loaded in the runtime environment, and may further include
information specifically describing the manner in which such files
were modified. For example, if the appliance is configured to
employ the SSH network protocol to provide encryption over secure
channels, one or more authentication keys may be automatically
generated when the appliance initially boots. As such, the runtime
logs displayed in operation 460 would include the automatically
generated encryption keys, which could then be used to
appropriately configure a network management system where the
appliance will eventually be deployed. Furthermore, the runtime
logs may include any files that were unused or underused subsequent
to the image being loaded in the runtime environment, and may
further include information specifically describing the manner in
which such files were unused or underused. For example, if one or
more files were not referenced or otherwise used when the image was
executed, the runtime logs may identify such files and mark the
unused files for deletion (e.g., to reduce a footprint for the
appliance). In addition, the runtime logs may further include a
package-level breakdown for unused files, wherein a percentage,
ratio, or other metric may be used to represent utilization levels
for specific packages. For example, the runtime logs may indicate
that ninety percent of the files included in a first package were
referenced or otherwise used when the image was executed, while
only ten percent of the files in a second package were referenced
or otherwise used (e.g., wherein the second package may be
recommended for removal from the appliance).
[0122] In one implementation, the user may interact with the
runtime logs and determine whether or not to modify the appliance
in view of any captured snapshots and/or any activity recorded in
the runtime logs. As such, in response to an operation 470
determining that the user has provided a request to modify the
appliance, a modified version of the appliance may be built in
operation 480. For example, the modified version of the appliance
built in operation 480 may include an image encapsulating one or
more of the snapshots captured in operation 440, wherein the
snapshots may be loaded to deploy the appliance in a specific state
corresponding to one or more of the snapshots. Furthermore, in one
implementation, operation 460 may include displaying the runtime
logs in a selectable manner, such that the user can precisely view
any modifications that occurred during execution (e.g., lines that
were added to, deleted from, and/or otherwise modified in a
particular file), in addition to an impact of the modifications
(e.g., an amount of data associated with various unutilized and/or
underutilized packages). As such, the user may review the runtime
logs and individually select one or more changes to apply in the
modified version of the appliance to be built in operation 480. For
example, a user may change a keyboard layout from English to
French/Canadian and select the particular changes in the runtime
logs for the change in keyboard layout to build a modified
appliance configured with a French/Canadian keyboard layout. Thus,
the user may view the runtime logs and individually select (or
unselect) any of the file modifications, unutilized files, and/or
underutilized files, wherein the modified appliance may be built in
response to the selected (or unselected) changes.
[0123] In one implementation, after the user has suitably executed
the image and/or built one or more images corresponding to modified
versions of the appliance, an operation 490 may include enabling
deployment for the appliance to any suitable runtime environment.
Alternatively (or additionally), the user may provide a request to
re-deploy the appliance in the hosted runtime environment, in which
case the method illustrated in FIG. 4 may be repeated, beginning
with operation 410. Thus, operation 490 may include enabling
deployment for any appliances that the user has created in and/or
uploaded to the virtualization environment, appliances published in
the appliance marketplace, appliances corresponding to certain JeOS
templates, and/or any other suitable appliances.
[0124] For example, in one implementation, the user may download
one or more images from the virtualization environment to a client
device for execution in an appropriate runtime environment at the
client device. Alternatively, in one implementation, the user may
load the image for execution at the client device using
bootstrapped network installation. For example, the image may
include a Preboot execution Environment (PXE), which may be
configured to bootstrap the client device and load the image in the
runtime environment (i.e., without the image being downloaded to
the client device).
[0125] Alternatively, instead of downloading or otherwise executing
the image at the client device (e.g., because of bandwidth
constraints), the user may deploy one or more images to a runtime
environment on a host web server or another suitable processing
resource where the image can be hosted and executed in an
appropriate runtime environment. Additionally, in one
implementation, one or more images may be deployed directly to a
cloud computing environment, such as an Amazon Elastic Compute
Cloud (EC2), wherein the deployed images may be executed in the
cloud computing environment. For example, the Amazon EC2
environment is generally provided as a commercial web service in
which users can create virtual server instances to deploy hosted
applications in a scalable manner. Thus, because Amazon EC2
provides a cloud computing environment using Xen virtual machine
instances that function as virtual private servers, deploying a
particular appliance to an Amazon EC2 cloud computing environment
may include building a Xen virtual machine image for the appliance,
which can then be deployed directly to the cloud computing
environment. However, although the example provided above describes
deployment to an Amazon EC2 cloud computing environment, the
appliance may be similarly deployed to any other hosted environment
and/or cloud computing environment capable of loading and executing
images.
[0126] According to one aspect of the invention, FIG. 5 illustrates
a flow diagram of an exemplary method for managing updates for a
virtual appliance. In particular, the method illustrated in FIG. 5
may begin in an operation 510, which may include detecting one or
more upstream updates to one or more origin repositories. For
example, a user may generally install various packages, patterns,
or other software components to be included in an appliance,
wherein the selected software components may be installed from one
or more origin repositories. Thus, as noted above in connection
with FIGS. 1 and 2, a repository metadata server may be configured
to detect upstream updates to one or more of the origin
repositories (e.g., when a local digest for a particular origin
repository does not match a current digest for the origin
repository). As such, in response to the repository metadata server
detecting an upstream update to one or more of the origin
repositories, operation 510 may further include an update server
identifying any particular packages, patterns, or other software
components associated with the upstream updates to the one or more
updated origin repositories.
[0127] In one implementation, the update server may then further
identify one or more appliances managed in the virtualization
environment that include the particular packages, patterns, or
other software components associated with the upstream updates. The
update server may then generate a workflow for updating the
identified appliances in an operation 520. For example, in one
implementation, the update workflow may generally include sending a
notification to a user that created a particular appliance having
one or more of the updated software components, wherein the
notification may indicate that updated versions have been released
for one or more of the software components installed for the
appliance. As such, in operation 520, the workflow may include
rebuilding an image for the appliance to install the updated
software components.
[0128] In one implementation, in response to the updated software
components being installed in the rebuilt image for the appliance,
the workflow may further include managing an integration test in an
operation 530. In particular, the integration test may generally be
performed to ensure that the rebuilt image functions or otherwise
behaves correctly with the updated software components installed.
For example, the updated software components may have different
features or behaviors from the original versions, which may not
necessarily integrate with other components installed for the
appliance. Accordingly, an operation 540 may include determining
verifying whether the rebuilt appliance functions or otherwise
behaves correctly with the updated software components
installed.
[0129] In one implementation, in response to operation 540
determining that the rebuilt appliance does not function correctly
with the updated software components installed, the workflow may
further enable the user to implement any changes modifying the
appliance in operation 550. For example, the user may implement the
changes within the integrated runtime environment in a similar
manner as discussed above in connection with FIG. 4, or the user
may alternatively employ one or more features of the development
environment described in further detail above to modify the
appliance. Thus, in operation 550, the user may implement the
changes to modify the appliance in any suitable manner to
accommodate the updated software components, wherein the modified
appliance may be rebuilt to incorporate the changes in operation
520 and subject to another integration test in operation 530. As
such, operations 520 through 550 may generally include rebuilding
the appliance and performing one or more integration tests in an
iterative manner until the appliance functions correctly with the
updated software components installed, as determined in operation
540.
[0130] In one implementation, in response to operation 540
determining that the rebuilt appliance functions correctly with the
updated software components installed, the rebuilt appliance
incorporating the updated software components may be released to
one or more users. For example, in one implementation, an operation
560 may include the update server determining whether one or more
downstream appliances exist for the rebuilt appliance that was
verified in operation 540 (e.g., the downstream appliances may
correspond to earlier versions of the rebuilt appliance, appliances
derived from the rebuilt appliance, etc.). In response to the
update server determining that one or more downstream appliances
exist for the rebuilt appliance in operation 560, the update server
may generate one or more updates corresponding to the rebuilt
appliance in an operation 570. As such, the updates may then be
released to earlier versions of the rebuilt appliance, wherein the
updates may be configured to install the updated software
components for the earlier versions of the rebuilt appliance
[0131] For example, in one implementation, the updates may be
bundled into a package or post-installation script that can be
distributed and installed to the earlier versions of the rebuilt
appliance, wherein the earlier versions may include appliances that
were cloned and/or derived to create the rebuilt appliance (e.g.,
as represented in a family tree constructed for the rebuilt
appliance). For example, the earlier versions of the rebuilt
appliance may periodically contact the virtualization environment
to determine whether any updates are available, and the update
server may release the updates generated in operation 570 to the
earlier versions of the rebuilt appliance. Furthermore, the update
server may be configured to provide access control over releasing
the updates to the earlier versions of the rebuilt appliances
(e.g., a particular update may be released to certain users, groups
of users, any users, etc.), and further to enable users of the
earlier versions of the rebuilt appliance to control the manner in
which the updates are released (e.g., the users may select or
deselect individual updates on a one-time basis, automatically
download all available updates, etc.).
[0132] In one implementation, in response to successfully
generating the updates in operation 570, or alternatively in
response to operation 560 determining that no downstream appliances
exist for the rebuilt appliance, an operation 580 may include
releasing the rebuilt appliance to subsequent users. For example,
releasing the rebuilt appliance may generally include replacing an
original version of the appliance with the appliance that was
rebuilt, tested, and verified in operations 520 through 550,
whereby subsequent users that clone, derive, or otherwise deploy
the rebuilt appliance may receive current versions for all of the
installed software components.
[0133] According to one aspect of the invention, FIG. 6 illustrates
a flow diagram of an exemplary method for managing support for a
virtual appliance. In particular, the method illustrated in FIG. 6
and described herein may be performed using a support engine having
a support analysis manager, which may be configured to analyze
supportability and service support requests for one or more
appliances managed in a virtualization environment. For example, as
described in further above in connection with FIG. 1, the
virtualization environment may provide various JeOS templates that
may be cloned to create base appliances that provide starting
points for developing virtual appliances, wherein one or more of
the JeOS templates may be based on an operating system distributed
by the provider of the virtualization environment. In addition, the
virtualization environment may further provide various packages,
patterns, and other software components that can be installed for
the appliances, wherein one or more of such packages, patterns, and
other software components may also be distributed by the provider
of the virtualization environment and/or trusted partners of the
virtualization environment provider.
[0134] Consequently, although an ISV, developer, or other user may
have knowledge relating to a particular application for which an
appliance has been customized, in many instances the ISV,
developer, or other user may lack detailed knowledge relating to
the operating system distribution and other software components
underlying the particular application for which the appliance has
been customized. As such, when customers or other users of the
appliance encounter problems with the custom appliance (e.g.,
because the appliance crashes or functions less than optimally),
the ISV, developer, or other user may lack sufficient knowledge to
provide support for the problems. Accordingly, in one
implementation, the virtualization environment may be configured to
enable users to submit one or more support requests in response to
problems that may be encountered with custom appliances, wherein
the support analysis manager may be configured to receive the
support requests in an operation 610, and further to service the
support requests using the techniques described in further detail
herein.
[0135] In one implementation, one or more of the support requests
received at the support analysis manager in operation 610 may
generally include a snapshot of the appliance for which support is
requested and a description of any problems with the appliance,
wherein the support analysis manager may then analyze the appliance
snapshot in an operation 620. For example, conventional techniques
for servicing support requests relating to problems with an
application often require a conversation between a user and a
technical support representative to resolve issues such as the
operating system being used by the customer, whether any service
packs are installed, whether certain caching features are enabled,
and/or various other issues relevant to diagnosing the problem in
order to provide the appropriate support. Furthermore, if the
support representative determines that the support requires the
application to be debugged, the support representative may ask the
customer to provide log files, interact with the application, or
perform other tasks. Accordingly, conventional techniques for
servicing support requests typically require a substantial amount
of interaction to resolve the problems, introducing inefficiency
and uncertainty to the support process.
[0136] As such, the method for servicing support requests
illustrated in FIG. 6 and described herein may include analyzing
the snapshot of the appliance in operation 620 to enhance the
information that can be referenced to provide the appropriate
support. For example, when a user experiences problems with an
appliance, the user may simply capture a snapshot of the appliance
while the appliance is in the problem state, such that the snapshot
analyzed in operation 620 may encapsulate an entire environment
associated with the problem state of the appliance. Thus,
diagnosing the problem state and providing the appropriate support
may be more efficient through the use of appliance snapshots,
wherein support representatives may simply interact with the
appliance snapshot to diagnose, debug, troubleshoot, or otherwise
provide the relevant support.
[0137] Additionally, in one implementation, the support analysis
manager may analyze the appliance snapshot in operation 620 to
determine whether or not the particular problem can be supported
(e.g., by the provider of the virtualization environment or a
trusted partner of the virtualization environment provider). In
particular, as noted above, in response to analyzing the appliance
snapshot in operation 620, the support analysis manager may
generate a support statement indicating whether support can be
provided for certain components installed for the appliance and/or
the particular problem described in the support request. For
example, in one implementation, the virtualization environment may
integrate a support analysis with a process for building the
appliance, wherein the integrated support analysis may include
ensuring that the appliance adheres to various supportability
guidelines, wherein the supportability guidelines may generally
describe certain changes or modifications that can or cannot be
implemented for one or more components supported by the provider of
the virtualization environment and/or trusted partners of the
virtualization environment provider.
[0138] In particular, during the build process, the support
analysis manager may analyze all of the packages, patterns, and
other software components installed for the appliance in view of
one or more heuristic rules to validate whether the appliance
adheres to the supportability guidelines (e.g., the heuristic rules
may specify permitted variants for packages, patterns, and other
software components available for installation in a particular
appliance). Thus, the support analysis manager may retrieve
metadata for the installed packages, patterns, and other software
components and analyze the metadata against the permitted variants
defined in the heuristic rules, which may express the
supportability guidelines in terms of the metadata. Accordingly, in
one implementation, the support analysis manager may analyze the
metadata for any supportable software components installed for the
appliance in view of any associated heuristic rules to generate the
support statement indicating whether or not the supportable
software components have been modified to the extent of being
rendered unsupportable.
[0139] Thus, in one implementation, the support analysis manager
may reference an original support statement generated when the
appliance was built in the virtualization environment, and may
further analyze the appliance snapshot to generate a current
support statement for the problem state associated with the
appliance. The support analysis manager may then determine whether
the provider of the virtual environment supports the problem
described in the support request in an operation 630. For example,
if the original support statement or the current support statement
indicates that the appliance has a modified kernel, new drivers, or
other customizations substantially altering an underlying operating
system distributed by the virtualization environment provider,
operation 630 may result in a determination that the virtualization
environment provider cannot support the problem. However, if one or
more of the support statements indicate that core components of the
base operating system remain intact, the determination in operation
630 may indicate that the virtualization environment provider
supports the problem. In such a case, an operation 640 may include
generating a response to the support request and delivering the
response to the user. For example, in one implementation, the
response may include a diagnosis of the problem, instructions for
resolving the problem, a modified version of the appliance that
resolves the problem, or any other suitable response. Furthermore,
in one implementation, the virtualization environment may be
configured to coordinate billing in response to the support request
being serviced (e.g., the virtualization environment provider may
bill users on a one-time basis, a subscription basis, an hourly
basis, an individual request basis, etc.).
[0140] In one implementation, in response to determining that the
virtualization environment provider cannot support the problem in
operation 630, an operation 650 may include determining whether
upstream support can be provided for the problem. For example, as
noted above, various packages, patterns, and other software
components available in the virtualization environment may
originate from trusted partners or other entities known to the
virtualization environment provider. Thus, in operation 650, the
support analysis manager may further determine whether the trusted
partners or other known entities can support the problem. For
example, the software components originating from the trusted
partners or other known entities may be associated with a
cryptographic key (e.g., a PGP certificate) that indicates whether
a particular package, pattern, or other software component is
"trusted," wherein the cryptographic key for the trusted package
may be configured to break or otherwise cease to be intact in
response to substantial alterations. Accordingly, operation 650 may
further include analyzing the original support statement and/or the
current support statement to further determine whether the problem
can be supported by the trusted partners and/or other known
entities, wherein the support request may be routed to the
appropriate supporting entity in an operation 660, thereby
coordinating support to resolve the problem for the user.
Furthermore, as described above, the virtualization environment may
be configured to manage billing for the support request in response
the support request being serviced or otherwise resolved (e.g., the
supporting entity may bill the virtualization environment provider,
which may bill the user in turn, or the virtualization environment
provider may bill the user and reimburse the supporting entity, or
the supporting entity may bill the user directly and reimburse the
virtualization environment provider, etc.).
[0141] In one implementation, in response to determining that the
virtualization environment provider, the trusted partners, and the
other known entities cannot support the problem in operations 630
and 650, the support request may be denied in an operation 670. For
example, if the problem relates to a custom component that the user
has installed for the appliance or components distributed by
third-parties not associated with the virtualization environment,
operation 670 may deny the support request because the problem does
not relate to components distributed by the virtualization
environment provider, the trusted partners, or the other known
entities. Furthermore, even if the problem relates to a component
distributed by the virtualization environment provider, the trusted
partners, or the other known entities, the support request may be
denied in operation 670 if the relevant component has been changed
or otherwise modified to the extent of being rendered
unsupportable. However, in one implementation, the support analysis
manager may further determine whether the change or other
modification to the relevant component can be resolved in a manner
that renders the problem supportable, wherein the denial of the
support request in operation 670 may include appropriate
instructions for making the problem supportable. For example, if
the problem is unsupportable because a particular package installed
for the appliance alters functionality or behavior of the operating
system, the denial of the support request may include instructions
indicating that the problem may be supportable in response to
uninstalling the package.
[0142] In one implementation, in response to appropriately
servicing the support request (i.e., by either responding to the
support request in operation 640, routing the support request in
operation 660, or denying the support request in operation 670), an
operation 680 may include providing feedback to the support engine.
For example, the feedback may include information relating to
whether the virtualization environment provider, the trusted
partners, and/or the other known entities successfully or
unsuccessfully resolved a supported problem, wherein such
information may be referenced to resolve similar problems that
arise subsequently, modify the heuristic rules or supportability
guidelines, or otherwise add to a knowledge base that the support
engine employs to analyze supportability and/or service support
requests. In another example, the feedback may include information
relating to problems for which support was denied, wherein such
information may be further referenced to determine whether or not
the heuristic rules or supportability guidelines should be
modified. For example, if a substantial number of users experience
similar problems with a particular unsupported package, the
virtualization environment provider may determine that support for
the package should be added, whereby developers, technical support
representatives, or other appropriate entities may be enlisted to
provide support for the package, and further to create appropriate
heuristic rules and supportability guidelines for the package.
[0143] Implementations of the invention may be made in hardware,
firmware, software, or various combinations thereof. The invention
may also be implemented as instructions stored on a
machine-readable medium, which may be read and executed using one
or more processing devices. In one implementation, the
machine-readable medium may include various mechanisms for storing
and/or transmitting information in a form that can be read by a
machine (e.g., a computing device). For example, a machine-readable
storage medium may include read only memory, random access memory,
magnetic disk storage media, optical storage media, flash memory
devices, and other media for storing information, and a
machine-readable transmission media may include forms of propagated
signals, including carrier waves, infrared signals, digital
signals, and other media for transmitting information. While
firmware, software, routines, or instructions may be described in
the above disclosure in terms of specific exemplary aspects and
implementations performing certain actions, it will be apparent
that such descriptions are merely for the sake of convenience and
that such actions in fact result from computing devices, processing
devices, processors, controllers, or other devices or machines
executing the firmware, software, routines, or instructions.
[0144] Furthermore, aspects and implementations may be described in
the above disclosure as including particular features, structures,
or characteristics, but it will be apparent that every aspect or
implementation may or may not necessarily include the particular
features, structures, or characteristics. Further, where particular
features, structures, or characteristics have been described in
connection with a specific aspect or implementation, it will be
understood that such features, structures, or characteristics may
be included with other aspects or implementations, whether or not
explicitly described. Thus, various changes and modifications may
be made to the preceding disclosure without departing from the
scope or spirit of the invention, and the specification and
drawings should therefore be regarded as exemplary only, with the
scope of the invention determined solely by the appended
claims.
* * * * *